2010 Subpart-W RTC Part2

VOLUME 8: COMPLIANCE AND ENFORCEMENT
8.0
COMPLIANCE AND ENFORCEMENT
No Comments Received.
8.1
COMPLIANCE ASSISTANCE
Comment Number: EPA-HQ-OAR-2009-0923-1011-4

Organization: El Paso Corporation
Commenter: Fiji George
Comment Excerpt Text:
IV. Proposed Rule Section 98.231 Applicability Analysis.(KEY ISSUE) EPA must provide
flexibility to regulated facilities with respect to initial year applicability analysis. El Paso
requests the EPA consider two screening methods to assist with the initial year applicability
determination. This proposal differs from the INGAA proposal. In addition, El Paso requests that
EPA promulgate a safe harbor provision for the first year of reporting. EPA can draw from prior
regulatory precedent granted by the Federal Energy Regulatory Commission (FERC). The FERC
granted a one-year safe harbor provision when it issued Order No. 704, which requires the
reporting of natural gas transactions (FERC Form No. 552). This provision gave the regulated
community the benefit of a rebuttable presumption that the data provided was accurate and
submitted in good faith and provided assurance that the regulatory agency (the FERC) would not
impose penalties for errors in reporting.
Response: For a response to this comment, please see the response to EPA-HQ-OAR-2009
0923-1011-30.
Comment: Section 98.231 Applicability Analysis: EPA must provide flexibility to regulated
facilities with respect to applicability determination of Subpart W rule. El Paso proposes two
screening methods to assist with determination.
EPA has not proposed a screening method to determine initial applicability to the rule but
instead relies on EPA proposed monitoring methods to make the determination of applicability to
the rule (per Subpart A provisions). The lack of a proposed screening method is a significant gap
with respect to an important aspect of compliance with Subpart W for entities that do not
currently have data collected using EPAs proposed methods (which we determined, will be most
entities). El Paso proposes two alternative screening methods:
1) A 10,000 metric tons CO2e threshold established based on combustion emissions using
procedures outlined in Subpart C; or
919
2) Allow use of best available data to estimate GHG emissions and determine applicability
against 25,000 metric tons CO2e threshold.
In addition, we request that EPA include a safe-harbor provision under both cases.
A company would have to conduct proposed monitoring at all its facilities to assess applicability
to the rule. Our current estimates of 156 facilities potentially subject to Subpart W are based on
the use of best available methodologies gained through our participation in the California
Climate Action Registry (CCAR) and Department of Energy 1605(b) emissions inventory
programs.
Without a screening method to assess initial applicability, a company like El Paso with potential
of over 150 applicable facilities operating in multiple states is subject to significant uncertainty
related to the actual applicability of the rule to its facilities. Such uncertainties will result in
significant compliance burden and hamper effective compliance planning.
A. Screening at 10,000 metric tons threshold based on Subpart C emissions:
We recommend that EPA permanently establish a screening threshold of 10,000 metric tons
CO2e per year based on emissions from combustion. That is, any facility with combustion
emissions (Subpart C) less than 10,000 metric tons would not be subject to Subpart W. We feel
this initial screening threshold is justifiable since repeated inventories conducted by the EPA
indicate methane emissions are approximately half of the total emissions from the natural gas
sector, and 10,000 metric tons is less than half the 25,000 metric tons threshold for the
monitoring and reporting rule. EPA estimates over 351 million metric tons 80 from facilities
subject to the proposed rule and over 415 million metric tons 81 from the entire sector. While
these estimates are almost double of EPAs official inventory 82 presented below in Table 3, the
breakdown between methane and CO2 related emissions remain at about 40% to 60% for the
overall natural gas sector. Therefore, a 10,000 metric tons threshold based on combustion
emissions is justifiable and provides adequate buffer considering the reporting threshold of
25,000 metric tons. Such a screening threshold will enable companies to focus on potential
major sources and not divert significant resources to confirming applicability or non
80
Preamble Page 18612
81
Economic Impact Analysis for the Mandatory Reporting of Greenhouse Gas Emissions under Subpart W
Supplemental Rule (EPA HQOAR20090923). Table 4-6.
82
EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2008.

920
applicability of smaller facilities.
This approach of basing initial screening on a 10,000 metric tons Subpart C threshold is El
Pasos preferred approach in that it presents a relatively clear, easily enforceable screening tool.
In the alternative, EPA could consider a screening approach based on best available data, as
outlined below. The 10,000 metric tons Subpart C screening method is, however, El Pasos
preferred approach due to its clarity and ease of application.
B. Screening based on best available data:
In the alternative, until monitoring has been conducted at a facility, EPA should allow potentially
affected Subpart W sources to employ best available data at a facility or onshore production
reporting area level to estimate GHG emissions and determine applicability against the 25,000
metric tons threshold based on the estimated emissions. The best available data includes 2006
IPCC Guidelines, U.S. GHG Inventory, DOE 1605(b), The Climate Registry, California Climate
Action Reserve and corporate industry protocols developed by the American Petroleum Institute,
the Interstate Natural Gas Association of America, and the American Gas Association and
measurement data developed internally by the reporters.
Companies should be allowed to use these best available data to estimate total emissions from
facilities or onshore production reporting areas for which monitoring at Subpart W facilities has
not yet been conducted, either because the rule is new or because the facilities have not yet been
estimated to exceed 25,000 metric tons based on best available methods. Companies should be
permitted to employ the estimated results to decide whether those facilities that have emissions
over 25,000 metric tons. Once a facilitys estimated emissions based on best available methods
exceed 25,000 metric tons, the first year monitoring methods would be carried out per the final
rule at these screened facilities and reports for the first year will be from these facilities that
have been screened to emit over 25,000 metric tons CO2e.
921
C. Safe harbor provision for screening methods:

Finally, El Paso requests a safe harbor provision be added under the final Subpart W rules. As
explained above, the proposed methods will employ screening techniques for applicability
determination in the first year. El Paso requests that EPA promulgate a safe harbor provision for
companies using any reasonable screening techniques. EPA can draw from such prior regulatory
precedence granted by the Federal Energy Regulatory Commission (FERC). The FERC granted
a one-year safe harbor provision after issuing Order No. 704, which requires the reporting of
natural gas transactions (FERC Form No. 552). This allowed the respondents to benefit from a
rebuttable presumption that the data provided is accurate and submitted in good faith. Further,
we [the FERC] do not intend to penalize respondents for errors in reporting on Form No. 552
provided that respondents use reasonable efforts to comply with the regulations regarding and
instructions for Form No. 552. We [the FERC] emphasize that the Commission expects
respondents submitting Form No. 552 in 2009 to do so in good faith and on a timely basis.
Such safe harbor provision will provide the necessary certainty for companies while limiting the
risk of potential retroactive enforcement in the event a facility initially determined not to be
applicable to Subpart W based on the screening methods, results with emissions > 25,000 metric
tons CO2e using EPAs final monitoring and estimation methods.
Response: EPA has reviewed the comment, and agrees that a screening tool should be
developed to serve as a guide to determining applicability under todays final rule. The threshold
for todays final rule is 25,000 metric ton CO2e, a threshold supported by many industry
stakeholders since it sufficiently captures the majority of GHG emissions in the United States,
while excluding most of the smaller facilities and sources. To facilitate the screening, similar to
what the Agency has already provided for other subparts of the Mandatory Reporting Program to
help reporters assess their applicability to the GHG reporting program, EPA plans to develop
voluntary screening tools for each petroleum and natural gas industry segment. EPA anticipates
that such tools would be based on easily determined inputs such as major equipment or
operational counts. The tools would be a guide to determine those facilities that are well below
the reporting threshold, those above, and those close to the threshold who will need to collect
further data to make a proper determination. EPA plans to place the screening tool for subpart W
on the following website:
http://www.epa.gov/climatechange/emissions/GHG-calculator/index.html.
Finally, with regard to the request for a safe harbor provision in using the screening tools, while
the tools would be designed to provide help to potential reporters for complying with the rule,
compliance with all Federal, State, and Local laws and regulations remain the sole responsibility
of each facility owner or operator subject to those laws and regulations. As regards to the safe
harbor provision in monitoring and collecting relevant data to report, EPA has in todays final
rule provided monitoring methods that are in most cases already in use in the industry, including
the use of emissions factors across several emissions sources. Therefore, EPA does not see any
need for providing a safe harbor. However, EPA has determined that for specified emissions
sources for certain industry segments, some reporters may need more time to comply with the
monitoring and QA/QC requirements of subpart W. In such cases, EPA concluded that
providing best available monitoring methods is reasonable where required for a specific period
922
of reporting. EPA has extensively detailed when and how reporters may use best available
monitoring methods in Section II.F of the preamble to todays final rule. Given the BAMM
provisions, EPA does not deem it necessary to provide safe harbor for the monitoring methods in
todays final rule. Finally, in regard to safe harbor for errors in reporting, please see response to
EPA-HQ-OAR-2008-0508-0952-1, excerpt 54.

Organization: Noble Energy, Inc
Commenter: Brian K. Lockard
Comment: The Proposed Rule essentially requires that Subpart W emission estimation methods
(monitoring and direct measurement) be applied to every industry segment-specific source
within an onshore petroleum and natural gas production reporting area every year. A screening
method that provides reasonable compliance certainty is needed to avoid unnecessary
compliance risk, implementation complexity, and financial burden.
Noble strongly recommends that a streamlined applicability screening method be included in the
rule for natural gas sector sources to preclude the need for monitoring and measurement in
reporting areas that fall below the applicability threshold. By defining an appropriate screening
method and conservative screening emission threshold to identify affected reporting areas,
compliance certainty can be assured and unnecessary measurement and monitoring can be
avoided.
Noble believes a first tier screening estimate for onshore petroleum and natural gas production
using a combination of API compendium emission estimation methods, and Natural Gas STAR
and area specific emission factors with a threshold of 20,000 tonne CO2e per year is a reasonable
screening approach. This approach would provide small reporting areas with relief from the
extensive emission calculation methods, and provide compliance and reporting certainty.
Noble offers it assistance to EPA for future industry studies and data collection to refine
screening tool(s) that will ensure reporting certainty for onshore petroleum and natural gas
production owners and operators.
Response: EPA agrees with the commenter that a screening tool would assist reporters in
threshold determination. EPA plans to develop voluntary screening tools for each petroleum and
natural gas industry segment, which will be web-based as is consistent for other EPA
rulemakings. For more details, please see response to EPA-HQ-OAR-2009-0923-1011-30.
Organization: Oklahoma Independent Petroleum Association
Commenter: Burck Halter
Comment: The proposed rule does not provide a simplified method for oil and gas businesses
operating production sites to determine applicability. EPA needs to develop screening tools to
help oil and gas operators determine applicability without the need to hire consultants. The
923
applicability tools need to be well defined and easy to understand so that all crude oil and natural
gas operators can easily apply it to their production sites. For example, EPA needs to establish
production thresholds/throughputs such as barrels of oil per day and thousands of cubic feet of
gas per day at which productions sites are exempt from further burdensome and costly data
collection. This would be similar to Subpart C of The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98) establishing an aggregate maximum rated heat input capacity
of stationary fuel combustion units at a facility of less than 30 mmBtu/hr. This will also make it
easy for operators to show compliance if audited/inspected by regulators.
Response: EPA agrees with the commenter that a screening tool would assist reporters in
threshold determination. EPA plans to develop voluntary screening tools for each petroleum and
natural gas industry segment. For more details, please see response to EPA-HQ-OAR-2009
0923-1011-30.
EPA has also in todays final rule established reporting thresholds for applicable emissions
sources. For more details, please see Section II.E of the preamble to todays final rule.
924
8.2
ROLE OF STATES

Organization: MidAmerican Energy Holdings Company
Commenter:
Comment: Many of the states in which MidAmerican subsidiaries operate are already subject to
state mandatory greenhouse gas reporting requirements; additional state and regional reporting
requirements are, likewise, expected to be implemented within the next two years. The potential
differences between state, regional and federal requirements relative to emission calculations and
scope of coverage will create significant burdens on reporting entities. MidAmerican encourages
EPA to coordinate with state and regional greenhouse gas reporting programs to develop a
common platform for reporting emissions.
Response: In developing todays final rule, EPA had evaluated reporting programs at the state,
regional, and federal levels. For the outcome of this evaluation, please see Sections A.4.a and
A.4.b of the Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background TSD (EPA-HQ-OAR-2009-0923-0027). The intended purpose of this evaluation
was to identify whether there was already a suitable monitoring and reporting program in place
of the requirements proposed under todays final rule and to locate emissions sources that are
subject to reporting under various monitoring and reporting programs. In short, prior to release
of the initial rule 40 CFR part 98, subpart W during 2009, EPA had already conducted significant
research into and evaluation of state and other programs. In order to develop an inventory
assessment based on a consistent set of assumptions, and control reporting burden, EPA was
required to develop a platform that provided the best and most consistent data from all segments
of the industry. EPA aknowledges that subpart W might not be identical to state and other data
collection activities currently in place. Different reporting programs have different goals and
objectives, and therefore different reporting requirements. Nevertheless, EPA has been working
with states to ensure that data can be exchanged between EPA and state level reporting
programs, thus facilitating the reduction of burden for reporters subject to multiple programs.
Please see response to EPA-HQ-OAR-2009-0923-1015-7 for further details.

Organization: Clean Air Task Force et. al.
Commenter: Pamela Campos
Comment: State and EPA Collaboration
Some states and jurisdictions may seek to coordinate their emissions reporting requirements with
EPAs data collection effort. We support effective collaboration between EPA and the states, and
encourage EPA to continue working cooperatively with WCI and individual states to enable
effective coordination to ensure broad public access to the most rigorous emissions data
available.
925
Response: EPA appreciates this comment, and has worked to coordinate with states,
recognizing that different programs have different goals. EPA has discussed subpart W with
states, and entities representing multiple states on data exchange, and will continue to work with
states so that the value of the MRR program can be utilized by states as best possible.
926
8.3
ENFORCEMENT

Organization: FLIR Systems, Inc.
Commenter: Thomas J. Scanlon
Comment: Penalties for Finding Gas Leaks
Some of our customers have confidentially expressed concern that the use of OGI equipment
could result in severe penalties since it is highly likely that the technology will detect gas leaks
that would never be identified with OVA or TVA technology. Through our field work and
training activities, we note that large leaks can occur over the course of time in natural gas and
petroleum systems through no negligence of the operator. To address concerns over the
compliance risks created by OGI requirements in Subpart W, EPA could consider amending
existing federal regulations (such as New Source Performance Standards for VOCs) to allow a
grace period for the correction of newly detected leaks at a facility that begins using OGI
technology. Texas has adopted this approach, in a bill that will take effect at the end of this
month. 83 This bill encouraged the adoption of OGI technology in a way that avoided unnecessary
resistance from the petroleum industry.
Response: EPA has considered comments received regarding monitoring methods in April 2010
proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002). In todays final rule, EPA is
permitting the use of alternative methods to OGI equipment for detection of equipment leaks for
certain emissions sources. For further elaboration on this topic, please see Section II.E in the
preamble of todays final rule. Amendment of New Source Performance Standards (NSPS) or
other regulations is beyond the scope of this rulemaking along with enforcement response which
is a case-by-case decision based upon a particular set of facts and circumstances. In general, EPA
attempts wherever possible to coordinate across rulemakings and programs.
Organization: Texas Oil and Gas Association
Commenter: Deb Hastings
Comment: Implement a phased approach of the rule over a period of several years.
Response: EPA has reviewed the comment and disagrees that a phased approach to the
implementation of todays final rule is required. For further details, please see response to EPA
HQ-OAR-2009-0923-1100-1, and a discussion of Best Available Monitoring methods in Section
II.F in the preamble of todays final rule.
83
House Bill (HB) 1526 of the 80th Legislature (2007), codified in Texas Health and Safety Code (THSC),
382.401, and in Texas Water Code, 5.752(2).
927
VOLUME 9: LEGAL ISSUES
9.0
LEGAL ISSUES

Organization: Kinder Morgan Energy Partners, L.P.
Commenter: Kim Dang
Responsibility for Reporting From Onshore Production Facilities. The proposed Subpart W
overlooks significant complications associated with determining reporting responsibility for
onshore production wells. Onshore production wells have extraordinarily diverse ownership and
operating arrangements. A single well can be owned by one entity, be operated by another entity,
lease portable equipment from a third entity, and have that portable equipment operated by yet
another entity. In these situations, the entities that directly operate certain equipment are in the
best position to gather emissions data for that equipment, whereas other entities working at the
same well site have limited ability to verify that data. Yet the proposed rule places the burden of
reporting entirely on the owner of the well or the holder of the operating permit. 84
Kinder Morgans operations place it in the position of being the owner of production wells,
lessor of equipment, lessee of equipment, and contract operator of production equipment. With
so many parties involved in reporting emissions from production equipment, Kinder Morgan is
concerned that it could be liable under the Mandatory Reporting Rule even when it has taken all
reasonable efforts for emissions information is properly reported to EPA. Thus, Kinder Morgan
respectfully requests that EPA provide the following clarification and safeguards with respect to
reporting responsibility in the onshore production sector:
a. Allow reasonable reliance on operators. Kinder Morgan is concerned that in cases where it is
responsible for reporting emissions from a particular well, and must therefore collect emissions
data from other entities operating equipment at that well, it could be held liable for non-obvious
errors or omissions committed by those other entities. The Mandatory Reporting Rule should
allow owners and permit-holders who are reporters to reasonably rely on data supplied by
operating companies associated with the production well. This safe harbor would not, of
course, apply where reliance is unreasonable, such as when the reporting entity knows that data
are erroneous or false.
b. Disallow vicarious liability for errors committed by reporting entity. Kinder Morgan is also
concerned that it may be held liable for errors or omissions committed by a reporting entity in
situations where Kinder Morgan is merely providing data to the actual owner or holder of the
operating permit at the well. EPA should clarify that an entity that provides properly collected
emissions data to a reporting entity will not be held liable if that reporting entity subsequently
commits an error or omission in reporting that data to EPA.
84
Proposed Subpart W, 75 Fed. Reg. at 18,614.

928
Response: EPA disagrees that Subpart W overlooks significant complications associated with
determining reporting responsibility for onshore production wells. The Agency has determined
that well owners or operators have control over rented, leased and contracted equipment and
services through the language in contracts, leases or other arrangements and while these may
require changes, ultimately the owner or operator is in the strongest position to obtain necessary
data.
The owner or operator through their designated representative (DR) is the entity that is
responsible for submitting the emissions data pursuant to todays final Rule. The DR may
provide in contracts, leases, or other agreements with third parties that true, accurate, and correct
reporting information must be provided to the DR in a timely fashion. If the third party fails to
provide timely, true, accurate, or correct information to the DR, then the DR has recourse
contractually, or otherwise, on the third party. While the DR or his delegates may need to
acquire necessary reporting information from a third party, the DR must make the appropriate
inquiries and certification when reporting; ultimate responsibility rests on him or her. Please see
Section II.F.5 of the preamble to todays final rule for a discussion of the designated
representatives responsibilities along with Section 98.4 of The Final Mandatory GHG Reporting
Rule (Final MRR), (40 CFR part 98).
The DR provisions are crafted similarly to the provisions of the Acid Rain Program (ARP), 40
CFR Part 72, and EPA has found that this approach provides a high degree of both data quality
and consistency and accountability. Similar comments were made about the data coming from
multiple owners and operators and the concerns about the certification of those data upon
promulgation of the ARP and the 2009 final GHG reporting rule to which EPA responded.
In recognition of the potential need to adjust contracts, leases, or agreements, additional
flexibility has been provided in todays final Rule to allow facilities to utilize best available
monitoring methods for a limited period. For further details, please refer to Section II.F of the
preamble in todays final rule. Moreover, to reduce burden, EPA has made provisions in todays
final rule to require external combustion equipment; that have a rated heat capacity less than or
equal to 5 mmBtu/hr to report equipment count by type; equipment above the threshold have to
report emissions using monitoring methods. See Section II.E of the preamble in todays final rule
for more information.
929
9.1
STATUTORY AUTHORITY
Topic 1: GHG Reporting under Subpart W and the Consolidated Appropriations Act
The following comments have been summarized to provide a common response;
Organization: Independent Petroleum Association of America
Commenter: Lee Fuller
Organization: Independent Petroleum Association of New Mexico
Commenter: Karin V. Foster
Comment Summary Text:
Commenters contend that EPAs authority to require GHG reporting stems exclusively from the
Consolidated Appropriations Act, 2008, that appropriations bills do not provide broad and
enduring authority for agencies to create new programs, are time limited, and create funding and
policy actions that cannot extend beyond the scope of the direction in the bill.
Some commenters also believe that the Appropriations Act upon which the rule is based violates
the separation of powers doctrine. As the statutory language of the Appropriations Act does
not, in its own terms, indicate what authority EPA is to invoke in order to require mandatory
emissions reporting of greenhouse gases or otherwise indicate a policy purpose for this
endeavor, Congress has asked EPA to establish a reporting program despite Congresss
unwillingness to date to itself enact a comprehensive GHG reporting and control regime. By
failing to delineate the policy behind, and boundaries of, this delegated authority, the
Appropriations Act violates the doctrine. Congress cannot delegate to an administrative agency
Congress law-making power or the power to exercise the discretion to modify a statute or power
to add parties or acts to those punishable under the statute.
Even were EPAs position that the rule is authorized by section 114 of the CAA correct, they
claim that the Subpart W proposal is in excess of EPAs statutory authority under CAA Section
114, fails to meet the standard of reasonableness applicable to requests for information under
Section 114, would fail to meet the arbitrary and capricious standard of review if challenged, and
EPA has exceeded its statutory authority under the Clean Air Act. They further claim that
930
Section 114 is the same section that authorizes EPA to issue administrative subpoenas to
individual companies to collect information (also known as Section 114 requests). To be valid
under Article 4 of the Constitution, subpoenas must be reasonable, relevant to the purpose, not
overly broad or unreasonably disruptive, and not unreasonably burdensome, and the recipient is
entitled to a finding as to whether EPA must pay the costs for responding to a burdensome
subpoena. The subpart they say fails to meet that standard.
They further claim that the rule lacks a reasonable relationship between the express statutory
purpose of collecting data reasonably necessary to inform GHG policy and the detailed and
onerous monitoring provisions it proposes to impose on oil and natural gas systems. In addition,
the rule is not supported by sound reasoning; fails to give reasonable consideration to important
aspects of the problem, including the effects and costs of the choices it proposes to make, and
fails to consider substantial arguments and alternative solutions.
Because GHGs are not yet subject to Section 110, 111, or 112, they also assert that Section 114
does not authorize EPA to promulgate the monitoring and reporting requirements of the
proposed rule. Section 114 can only be used for regulated pollutants and provides authority for
EPA to seek information for carrying out any provision of this chapter. CO2 is not, however,
subject to this chapter. The term chapter means Chapter 85 of Title 42, i.e. the Clean Air Act
Although the U.S. Supreme Court decided in Massachusetts v. EPA, 549 U.S. 497 (2007) that
CO2 is an air pollutant, until EPA makes a final endangerment finding and a control
requirement is established under the Act, CO2 is not a pollutant subject to the Act and thereby
not a pollutant subject to the Chapter. Nor can EPA rely on its authority to do studies and
investigations under Section 103 of the CAA to bootstrap into authority under Section 114 for
the proposed rule. Section 103 does not mention the words monitoring or reporting, only
words such as studies, surveys, investigations, and similar discrete, one-time or timelimited activities. When applied to the CAA, the rule of statutory construction, ejusdem generis,
indicates that EPAs statutory authority is limited. Section 821 of Public Law 101-549,
expressly authorizes EPA to require monitoring and reporting for CO2 on certain electric
generating units. EPAs authority under Section 821 is limited, however, to certain power
generating units subject to the provisions of the Acid Rain program. If EPA had the authority it
now claims to have under Section 114 (bootstrapped by Section 103), it would have been
unnecessary for Congress to enact Section 821 to establish ongoing CO2 monitoring and
reporting requirements for power generating units, not to mention the monitoring requirements in
Section 112, Title IV, and Title VI of the CAA, among others. A canon of statutory construction
is that one section of an Act should not be read so as to usurp or make another section
superfluous.
One commenter claims that use of Section 114 as the basis for the rule runs counter to its
longstanding use of the section in the past which as been limited to issuing Information
Collection Requests (ICRs) and that ICRs are limited to collecting data from specific sources
over a discrete period of time. Section 114 has not previously been used to justify an essentially
economy-wide, permanent rule that requires all affected sources to submit new data annually on
an ongoing basis and imposes new monitoring and measurement processes to collect such data.
EPAs claims that it has the authority to require all persons whom the Administrator believes
may have necessary information on emissions to report to the agency under CAA Section 114
931
and that the reporting requirement may even extend to persons not otherwise subject to CAA
requirements is a capricious extension of the legislative intent of the Clean Air Act.
Specifically, the Act has a list of entities under the jurisdiction of the CAA. To assume that it
also applies to all petroleum and natural gas facilities operators based on an Administrators
subjective beliefs will result in arbitrary bureaucratic requirements to report under this rule
Summary Response: Regarding the EPAs legal authority to establish reporting requirements,
the Appropriations Act as exclusive authority, Appropriations Act as violating the separation of
powers doctrine, lack of a reasonable relationship between express statutory purpose and
monitoring requirements, relationship between Section 114 and Sections 103, 110, 111 and 112,
and 821 please see the Section II.Q, Summary of Comments and Responses on Statutory
Authority, of The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98)
preamble, Volume 9, Legal Issues, of the EPAs Response to Public Comments for the Final
MRR and the response in Section I.C, Legal Authority, of the preamble to todays final rule.
Also see [Volume 10, Cost and Economic Impacts of the rule] and preamble Section III
respecting effects and costs of the choices proposed and alternative solutions generally. The
argument that Section 114 is EPAs subpoena power is incorrect. The Agencys authority to
issue subpoenas is in Section 307(a) and an attempt to engraft limitations of authority under
Section 114 by mischaracterization as a subpoena authority is misfounded. Nevertheless, EPAs
exercise of its authority under 114 is reasonable, relevant to the purpose, not overly broad nor
unreasonably disruptive. See, Volume 9. Some burden in response to Agency inquiries in
furtherance of its legitimate interests is not unreasonable.
Similarly, commenter confuses Section 114 and ICR requirements under the Paperwork
Reduction Act, 44 U.S.C. 3501, et seq. EPA has issued an ICR for this rulemaking, has been
assigned an EPA ICR number by OMB and has fully complied with PRA requirements. See,
preamble Section IV.B See also, Volume 9 explaining that while it might be generally true that
EPA has used its information gathering authority in a more targeted manner, nothing in Section
114 precludes this broader, yet still targeted rulemaking.
The claim that the Subpart W proposal is in excess of EPAs statutory authority under CAA 114,
fails to meet the standard of reasonableness applicable to requests for information under Section
114, would fail to meet the arbitrary and capricious standard of review if challenged, and that
EPA has exceeded its statutory authority under the Clean Air Act is conclusory and is not
otherwise specifically supported by the commenter. Without more, EPA cannot directly respond
other than to refer commenter to its position as its authority to use and to the reasonableness of
using Section 114 as a basis for the rule as set forth in Volume 9. Similarly, the claim that
EPAs exercise of its authority under Section 114 to include all petroleum and natural gas
facilities operators is a capricious extension of legislative intent a only a conclusion on the
commenters part without supporting rationale to which response cannot be formulated.
As referenced above and explained in Vol. 9, Section 114 can apply to pollutants even though
they might not be subject to regulation. Moreover, we note that EPA has taken final action that,
as of January 2, 2011, makes the air pollutant comprised of the mix of six greenhouse gases
subject to regulation under the Act. See, e.g., Reconsideration of Interpretation of Regulations
that Determine Pollutants Covered by Clean Air Act Permitting Programs; Final Rule, 75 FR
932
17004 (April 2, 2010) (explaining that greenhouse gases would become subject to regulation on
the takes effect date of the Light Duty Vehicle Rule (Light-Duty Vehicle Greenhouse Gas
Emission Standards and Corporate Average Fuel Economy Standards; Final Rule, 75 FR 31514)
which was published on May 7, 2010.
Organization: Colorado Department of Public Health and Environment
Commenter: Kirsten King
Disregards local agency, state or national boundaries, requiring emission sources to report
aggregated emissions per basin, which may involve multiple local, state and national authorities.
Response: The GHG Reporting Program was established to collect data on national levels of
greenhouse gas emissions by different sectors of the economy. For further information, please
see the preamble of The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part
98), Section I.E and the response to comment EPA-HQ-OAR-2009-0923-1044-1 for further
information. It is not possible to match every local or state boundary and still manage reporting
under a reasonable facility definition that applies to all operators across the country. Please see
response to EPA-HQ-OAR-2009-0923-1015-7 for further details.
Commenter: Karin Ritter
Organization: American Petroleum Institute, Karin Ritter
EPA Lacks the Legal Authority for the Reporting Rule and Proposed Subpart W
EPA asserts that it developed the proposed Subpart W reporting requirements based on existing
authority under Clean Air Act (CAA) Section 114. See 75 Fed. Reg. 18,610. EPA, however, is
over-reaching this authority, which does not authorize it to require the proposed indefinite and
sweeping monitoring, recordkeeping, and reporting from the facilities covered by proposed
Subpart W, regarding emissions with a currently uncertain regulatory status. At a minimum,
none of EPAs stated purposes for the data justify the frequency and duration of the reporting
requirements, or the imposition of burdensome new measurement protocols and the installation
of extensive and expensive instrumentation. Further, EPA already has in its possession and
continues to collect detailed GHG emissions inventory data that is sufficient to meet the stated
purposes the Agency asserts underlie the proposed rule.
i. EPAs Interpretation of Its Section 114 Authority is Overly Broad
In the preamble to proposed Subpart W, EPA suggests it is appropriate to gather information
required under the proposed rules because the information may be relevant to EPA
implementation of a variety of CAA provisions. The Agency asserts that CAA section 114
provides EPA broad authority to require the information proposed to be gathered by [the rules]
because such data would inform and are relevant to EPAs carrying out a wide variety of CAA
provisions. 75 Fed. Reg. 18,610. In the proposed Subpart W, EPA does not specify any
particular purpose of CAA provisions for which it needs this information. The Agency merely
933
indicates that it is comprehensively considering how to address climate change under the CAA,
including both regulatory and non-regulatory options. Id. at 18,611.
To begin, these generalized suppositions as to how the information might be used at some
indeterminate time in the future provide wholly inadequate justification for an information
request of the size, scope, and duration that EPA has proposed. EPA has a fundamental
obligation to assert a rational basis for implementing its authority under Section 114, which
includes in this case a particularized explanation of the reasons EPA actually, currently needs
this information (versus a wish list of programs and policies that might be informed by
gathering this information). In short, the proposal provides no clear, specific, and ascertainable
explanation of why this information is needed and how it will be used and, thus, the rule stands
to be an unreasonable and arbitrary exercise of Section 114 authority.
Moreover, the agencys reading of its authority under Section 114 is overly broad and would
improperly render meaningless the limitations in this provision. EPA emphasizes that
information may be required under Section 114 for purposes of carrying out any provision of
the Act. However, Section 114 has traditionally been limited to discrete information requests
from particular emission sources, and EPAs use of this provision for an ongoing reporting
program for entire sectors of the economy is unprecedented.
Section 114 must be read in light of the entire statute, which makes clear that the data collection
it authorizes is limited to only certain persons that are subject to the Acts requirements. 42
U.S.C. Section 7414(a)(1) (limiting Section 114 applicability to any person who owns or
operates any emission source, who manufactures emission control equipment or process
equipment, who the Administrator believes may have information necessary for the purposes set
forth in this subsection, or who is subject to any requirement of this chapter (except a provision
of subchapter II . . .)). The applicability of Section 114 is thus limited to entities who own or
operate an emission source or who are subject to regulation under the CAA for a given air
pollutant. EPAs attempt to collect data under Section 114 from persons who are not owners or
operators of an emission source or who are not subject to regulation under the CAA exceeds
Section 114s authority and is improper. It is a fundamental rule of statutory construction that
statutes should not be interpreted in a manner so as to make any phrase redundant. See, e.g.,
Gustafson v. Alloyd Co., Inc., 513 U.S. 561, 574 (1995) (the Court will avoid a reading which
renders some words altogether redundant); Zimmerman v. Cambridge Credit Counseling Corp.,
409 F.3d 473, 476 (1st Cir. 2005) (no construction should be adopted which would render
statutory words or phrases meaningless, redundant, or superfluous.); U.S. v. Hovsepian, 359
F.3d 1144, 1160 (9th Cir. 2004) (We interpret statutes so as to avoid making any phrase
meaningless or unnecessary.). EPAs interpretation of Section 114 ignores and violates this well
established canon by requiring indefinite data collection from any person based on the vague
stated purpose of carrying out any provision of the Act. EPA cannot expand such authority
beyond the express limitations of Section 114, and permit that single phrase to swallow the
express limitations of the entire rule.
ii. The 2008 Appropriations Act Does Not Provide Legal Authority for the Rule
Clearly, EPAs decision to propose Subpart W is motivated by the 2008 Consolidated
Appropriations Act, signed into law on December 26, 2007. Consolidated Appropriations Act,
934
2008, P.L. 110-161, 121 Stat. 1844, 2128 (2008) (Appropriations Act). This Act authorized
one-time funding for activities to develop and publish a draft rule not later than 9 months after
the date of enactment of this Act, and a final rule not later than 18 months after the date of
enactment of this Act, to require mandatory reporting of greenhouse gas emissions above
appropriate thresholds in all sectors of the economy of the United States. Id. The
straightforward point of the appropriations bill provision is to have EPA administer the
collection of data; Congress did not draw any connection (as EPA attempts to do in its preambles
to the two proposed rules) to carrying out any aspect of the CAA. Nor did the Appropriations Act
provide EPA any new authority under the CAA to promulgate mandatory GHG reporting rules.
Therefore, EPAs reliance on Section 114 is misplaced because, as explained above, these
Sections may only be invoked for specified purposes under the CAA. General information
gathering pursuant to the appropriations language is not within the scope of authority conferred
by Section 114.
Moreover, the narrowly prescribed activity funded by the Appropriations Act necessarily
constrains the scope of EPAs information gathering under Section 114. First, given the limited
funds authorized by Congress, it would be impossible to maintain the indefinite and overly
expansive reporting program for all sources in the onshore and offshore petroleum and natural
gas sectors proposed. Second, the Appropriations Act does not prescribe any enforcement
authority. Because the CAA does not independently provide the authority for the proposed rule,
it cannot be the basis for any enforcement action. Congressional grants of enforcement authority
must be explicit. Cf. Marshall v. Gibsons Products, Inc. of Plano, 584 F.2d 668, 675 (5th Cir.
1978) (Congress is cognizant of the need to set forth explicitly the authority of an administrator
or agency to seek enforcement relief in federal court.).
Response: For a response to this comment, please see EPAs Response to Public Comments,
Volume No. 9, Legal Issues, October 30, 2009., and II.Q of the preamble to The Final
Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), as well as Section II.F of
the preamble to todays final rule.
.
Topic 2: Aggregation of Gathering and Boosting Systems with Processing Facilities
The following comments have been summarized to provide a common response;
Organization: Latham & Watkins LLP
Commenter: Matthew C. Brewer
935


936

Organization: Aka Energy Group, LLC
Commenter: Barbara Wickman
Comment Summary Text:
The proposed GHG reporting rules super source aggregation of the natural gas processing
sector violates the Clean Air Acts stationary source requirements and otherwise is unlawful.
Natural gas processing plants have been regulated under CAA programs as discrete stationary
sources and have never been aggregated with gathering operations for purposes of applying
thresholds and other requirements.
Commenter claims that the proposed GHG reporting rule is inconsistent with EPAs legal
necessity findings for effective GHG regulation set forth in the tailoring rule. The doctrines of
absurd results, administrative necessity and one-step-at-a-time are intended to preserve a
rational major and minor source distinction for GHGs so that the GHG permitting can apply
to major sources as defined in the PSD program.
They also contend that the stationary source legal framework mandates a presumption against
aggregation because the Clean Air Act regulates sources and the various Clean Air Act
programs applying to stationary sources have either or both statutory and regulatory source
definitions that depend upon three key elements, to wit: (1) belong to same industrial group,
are (2) located on one or more contiguous or adjacent properties, and are (3) under common
control. super source aggregation is unprecedented in EPAs forty year history of regulating
the natural gas process sector where typically the various regulations under the CAA define a
facility as a group of emissions sources all located in a contiguous area and under the control of
the same person (or persons under common control).
In further support, they claim that in the 1990 CAA amendments, Congress endorsed EPAs nonaggregation source approach to the natural gas industry by expressly codifying it under the Title
III HAPs Program which approach, as expressed in Alabama Power Co. v. Costle, 636 F.2d 323
(DC Cir. 1979) over 30 years ago, and reflected in EPAs regulations and guidance, was
expressly codified by Congress for the natural gas industry in Section 112(n)(4) of the Act.
Their position is further supported, they state, two recent pronouncements addressing
aggregation in the natural gas industry: Memorandum from Assistant Administrator Gina
McCarthy re: Withdrawal of Source Determinations for Oil and Gas Industries (September 22,
2009) and Order Responding To Petitioners Request That The Administrator Object To
Issuance Of A State Operating Permit (October 8, 2009) instructing the Colorado Department of
Public Health and Environment, Air Pollution Control Division to reevaluate a decision it had
made with respect to a natural gas compressor station owned by Kerr McGee Corporation.
These positions indicate that only on a case-by-case basis may it be appropriate to aggregate
emissions from wells or compressors in a gathering system and thus provide no basis for the
proposed rules super source aggregation of an entire gathering system with natural gas
937
processing plant. In particular, the McCarthy memorandum reaffirmed that the three key factors
-- same industrial group, contiguous or adjacent location, and common control -- apply on a
case-by-case basis to determinations of whether aggregation of emissions in the natural gas
industry are appropriate; thus aggregation of natural gas processing systems as proposed is
inappropriate.
Section 114 allows information gathering to implement and enforce substantive regulatory
provisions only, and therefore, does not provide a legal basis for EPA to impose a unique
source definition that conflicts with its history of regulating the natural gas processing sector
they argue. Specifically, Section 114 authorizes information gathering solely for the purpose of
developing or assisting in the development of regulations under the provisions specifically
enumerated in Section 114, enforcement of existing regulations or carrying out any other
provision of the Act.
With respect to the Proposed Subpart W, Section 114 provides EPA no authority to implement
super source aggregation for the natural gas processing sector independent of and in direct
conflict with the regulatory history of the Acts application to the natural gas processing sector.
Nor has EPA, they aver, provided any rational explanation as required by section 114 for how
GHG emissions information gathered on a super source aggregation basis will support its
implementation of GHG regulatory programs for the natural gas processing sector. Application
of Subpart Ws reporting requirements to the sector as defined in the rule as proposed is not
reasonably limited in scope and time as required by various aspects of Section 114.
The proposed rule is arbitrary and capricious because epas reference to materially different,
non-clean air act reporting programs, such as DOT Pipeline and Hazardous Materials Safety
Administration (PHMSA) regulations and the Climate Registry provides no as justification for
super source aggregation commenter contends.
Finally, in support of its averment that the proposed natural gas processing facility definition is
legally flawed, commenter believes that the approach is simply unfair and inequitable. In taking
such an approach here, EPA would be setting a precedent that any industries that have a complex
manufacturing and distribution chain could be required to aggregate all emissions associated
with a product, including those from geographically distant facilities and emissions associated
with the transport of materials among facilities. If this is not the case, then the application of the
super source aggregation approach to only the natural gas processing sector is simply unfair. In
particular, the requirement that the natural gas processing sector report fugitive emissions from
gathering systems while not requiring similar reporting from the natural gas transmission sector
is fundamentally inequitable.
Summary Response: As is clear from the submission, these comments are peculiar to and
squarely relate to the onshore natural gas processing segment of subpart W and the inclusion of
gathering operations in the source category segment. In todays final rule, EPA has revised the
definition of the segment to eliminate reference to field gathering and/or boosting stations that
gather and process natural gas from multiple wellheads, and compress and transport natural gas
(including but not limited to flowlines or intra-facility gather lines or compressors) as feed to the
938
natural gas processing facilities as being a part of the processing facility, as well as reference to
gathering and boosting stations that send the natural gas to an onshore natural gas transmission
compression facility, or natural gas distribution facility, or to an end user being considered stand
alone natural gas processing facilities. Therefore, while EPA does not agree with the positions
taken in support of commenters argument, because gathering operations relating to natural gas
processing facilities are not included in the source category and the commenters concern that
EPA is unlawfully augmenting the source category by their inclusion is a moot point. As EPA
interprets the comments, they are limited exclusively to the onshore natural gas processing
segment making further response unnecessary. EPA may choose to cover emissions from this
portion of the industry at a later time by further rulemaking or otherwise. With respect to the
scope of EPAs authority under Section 114, see Volume 9, Legal Issues, of the EPAs Response
to Public Comments for the Final MRR.
9.1.1 SEGREGATE OFFSHORE ISSUES SEPARATELY

OCS Authority
EPA is proposing to amend 40 CFR 98.2(a) so that the Final Mandatory GHG Reporting Rule
applies to facilities located in the United States and the Outer Continental Shelf.
Additionally, EPA is proposing revisions to the definition of United States to clarify that it
includes the territorial seas. This ensures that the facilities located offshore of the United States
that are injecting CO2 into the sub-seabed for long-term containment will also be required to
report data regarding GHG emissions. EPA writes:
Together these changes make clear that the Mandatory GHG Reporting Rule applies to facilities
on land, in the territorial seas, or on or under the Outer Continental Shelf, of the United States,
and that otherwise meet the applicability criteria of the rule. 85
EPA notes that its Clean Air Act (CAA) authority to collect emission information from certain
offshore petroleum and natural gas platforms has been questioned:
Some commenters argued that EPA does not have the authority to collect emissions information
from offshore platforms located in areas of the Western Gulf because they are under the
jurisdiction of the Department of the Interior. They cited, among other things, the Outer
Continental Shelf Act, 43 U.S.C. 1334. Without opining on the accuracy of the commenters
85
Federal Register Vol. 75, No. 69, 18611.

939
summary of OCSLA or other law, we note that even the commenter describes these authorities as
relating to the regulation of air emissions. Todays proposal does not regulate GHG emissions;
rather it gathers information to inform EPAs evaluation of various CAA provisions. Moreover,
EPAs authority under CAA Section 114 is broad, and extends to any person who the
Administrator believes may have information necessary for the purposes of carrying out CAA,
even if that person is not subject to the CAA [emphasis added]. 86
We support EPAs inclusion of OCS sources in Subpart W. We strongly agree with the EPA that
it plainly has authority to collect emissions information from offshore oil and gas platforms. 87 As
the agency states, EPAs authority under [Clean Air Act] Section 114 is broad.[Id.] Indeed,
that section gives EPA authority over any person who owns or operates any emission source,
including the power to establish monitoring systems, require equipment to be installed, and
collect any other information as the Administrator may reasonably require. 88 Nothing in the
statute indicates this power stops at the beach, and such a limit would be entirely inconsistent
with the provisions purpose of providing high quality emissions data to the public, as offshore
platforms are a major pollution source.
EPA writes that some commenters, nonetheless, suggest that the Outer Continental Shelf Lands
Act (OCSLA), 43 U.S.C. 1331, somehow limits EPAs authority. These commenters have
the law backwards. First of all, nothing in OCSLA purports to limit EPAs authority, so the
commenters would have to argue that OCSLA implicitly repeals Section 114 authority over
offshore activities even though the statute does not contain a word of text doing so. That
argument runs foursquare into [the] presumption against implied repeals. 89 That powerful
presumption states that absent a clearly established congressional intention, repeals by
implication are not favored. An implied repeal will only be found where provisions in two
statutes are in irreconcilable conflict. 90 Nothing in OCSLA and Section 114 are irreconcilable,
so the agency, rightly, must presume Congress did not silently strip its powers away. Indeed,
OCSLA, if anything, reinforces EPAs authority. That statute explicitly extended the laws and
civil and political jurisdiction of the United States to the Outer Continental Shelf, cementing the
relevance of domestic environmental law to that region. 91 In doing so, Congress recognized that
the shelf is:a vital national resource reserve held by the Federal Government for the public, and
which is subject to environmental safeguards. 92 In fact, Congress even ordered the Secretary of
the Interior, who oversees leasing programs in the region, to cooperate with the relevant
departments and agencies of the Federal Government to enforce environmental laws. 93 In
86
Federal Register Vol. 75, No. 69, 18611.
87
See 75 Fed. Reg. at 18,611.
88
42 U.S.C. 7414(a) (emphasis added).
89
See National Assn of Home Builders v. Defenders of Wildlife, 551 U.S. 644, 665 (2007).
90
Lockhart v. United States, 546 U.S. 142, 149 (2005).
91
See 43 U.S.C. 1333(a)(1).
92
See 43 U.S.C. 1332(3).
93
See 43 U.S.C. 1334(a).

940
short, there is no serious argument that EPA lacks authority for its actions. If anything, EPA
would be remiss if it did not account for major offshore oil and gas emissions sources. Offshore
petroleum and natural gas systems GHG emissions must be included in EPAs overall national
assessment of all sectors of the economy of the United States, as required. Thorough data is
needed to develop a national policy that addresses GHG emissions as a whole, and not in a piece
meal fashion.
Response: EPA agrees with the commenter that it has jurisdiction to gather information from
offshore facilities. See also, Response to EPA-HQ-OAR-2009-0923-1151-16.
Organization: American Petroleum Institute
D. Offshore Petroleum and Natural Gas Production Facilities in the Western Gulf of Mexico
i. EPA Authority Over the Western and Central Gulf of Mexico
Under the proposed Subpart W, offshore petroleum and natural gas production facilities located
in both State administered waters and Mineral Management Services administered Federal
waters would be obligated to report GHG emissions. The platforms in the Western and Central
Gulf of Mexico should be excluded from the reporting rule because the MMS has jurisdiction for
air emissions from facilities on the Outer Continental Shelf (OCS).
Under the Outer Continental Shelf Lands Act (OCSLA), the Department of Interior (DOI)
has the authority to regulate air emissions on the OCS. U.S.C. Section 1334. In California v.
Kleppe, the court held that the OCSLAs specific grant of jurisdiction to DOI precluded the
application of Prevention of Significant Deterioration regulations to E&P facilities on the OCS.
604 F.2d 1187 (9th Cir. 1979). In so ruling, the court specifically rejected EPAs argument that
dual jurisdiction over air pollution on the OCS existed between EPA and DOI.
After Kleppe, the 1990 Clear Air Act Amendments established the authority of EPA to regulate
air pollution on the OCS from oil and gas exploration and production facilities (referred to as
OCS sources), but as to the Gulf of Mexico, limited EPAs air pollution jurisdiction over OCS
sources to the area of the Gulf of Mexico east of longitude 87 degrees and 30 minutes and
confirmed DOIs air pollution jurisdiction to the area of the Gulf of Mexico west of longitude 87
degrees and 30 minutes (Western and Central Gulf). Air pollution from OCS activities, 42
U.S.C. Section 7627, provides as follows:
(a)(1) Applicable requirements for certain areas
Not later than 12 months after November 15, 1990, following consultation with the Secretary of
the Interior and the Commandant of the United States Coast Guard, the Administrator, by rule,
shall establish requirements to control air pollution from Outer Continental Shelf sources located
offshore of the States along the Pacific, Arctic and Atlantic Coasts, and along the United States
Gulf Coast off the State of Florida eastward of longitude 87 degrees and 30 minutes (OCS
941
sources) to attain and maintain Federal and State ambient air quality standards and to comply
with the provisions of part C of subchapter I of this chapter. For such sources located within 25
miles of the seaward boundary of such States, such requirements shall be the same as would be
applicable if the source were located in the corresponding onshore area, and shall include, but
not be limited to, State and local requirements for emission controls, emission limitations,
offsets, permitting, monitoring, testing, and reporting The authority of this subsection shall
supersede section 5(a)(8) of the Outer Continental Shelf Lands Act [43 U.S.C.A. Section
1334(a)(8)] but shall not repeal or modify any other Federal, State, or local authorities with
respect to air quality
(b) Requirements for other offshore areas
For portions of the United States Gulf Coast Outer Continental Shelf that are adjacent to the
States not covered by subsection (a) of this section which are Texas, Louisiana, Mississippi, and
Alabama, the Secretary shall consult with the Administrator to assure coordination of air
pollution control regulation for Outer Continental Shelf emissions and emissions in adjacent
onshore areas. Concurrently with this obligation, the Secretary shall complete within 3 years of
November 15, 1990, a research study examining the impacts of emissions from Outer
Continental Shelf activities in such areas that fail to meet the national ambient air quality
standards for either ozone or nitrogen dioxide. Based on the results of this study, the Secretary
shall consult with the Administrator and determine if any additional actions are necessary.
In proposed Subpart W, EPA has expressly stated that the CAA is the source of its authority for
the proposed rule. Yet, there is no authority under the CAA for the collection of information with
respect to geographic areas or industry segments over which EPA has no jurisdiction, such as the
Western and Central Gulf. See 42 U.S.C. Section 7414(a) (enumerating authorized purposes for
EPAs seeking information from the regulated community). Instead of EPA jurisdiction over
OCS sources in the Western and Central Gulf, under the CAA, the Secretary of DOI is merely
required to consult with the Administrator of the EPA with respect to OCS emissions with
regards to the Western and Central Gulf, not vice versa.
Congress has recognized that the Western and Central Gulf of Mexico is a unique national
resource. In order to promote and regulate the expeditious and orderly development of this
resource, Congress made a conscious decision to place the regulation of air emissions from OCS
sources with DOI, an agency that has specialized expertise with regard to exploration and
production activities and their effect on both the OCS and onshore environments. Under
OCSLA, Congress has chosen DOI as the lead and exclusive agency for all aspects of air
emissions in the Western and Central Gulf. EPAs proposed rule undermines Congress
delegation of this authority to DOI.
Finally, EPA cannot justify applying this proposed rule to the Western and Central Gulf by
stating that it is under a directive to collect information from all industry segments. The general
statement that emissions from all segments should be reviewed cannot be fairly interpreted as an
expansion of jurisdiction where there previously was none. Nor can EPA justify this proposed
rule by stating that information collection is not regulation. This issue was addressed in the
Williams Companies v. FERC, 345 F.3d 91 (D.C. Cir. 2003). In Williams, the court found that
942
the FERC could not require companies to submit certain information relating to pipeline
transportation rates on OCS pipelines because Congress gave DOI (through OCSLA) authority to
regulate open and non-discriminatory access to OCS pipelines.
As applied to OCS sources in the Western and Central Gulf, the proposed rule is beyond EPA
authority, violates both the CAA and OCSLA (and possibly other federal statutes such as the
Administrative Procedures Act, the Paperwork Reduction Act, etc.), and is unfair, overly
burdensome and inefficient. For these reasons, the Western and Central Gulf should be excluded
from the proposed rule.
Response:
The commenter did not respond directly to EPAs statements in the proposed rule regarding the
agencys broad authority to collect information under section 114 of the CAA, including from
sources in the Western Gulf. Moreover, these comments reflect a misunderstanding of the scope
of EPAs authority under section 114 of the CAA. EPAs authority under CAA Section 114 is
broad, and extends to any person who the Administrator believes may have information
necessary for the purposes of carrying out the CAA, even if that person is not subject to the
CAA. Indeed, by specifically authorizing EPA to collect information both from persons subject
to any requirement of the CAA, and from any person who the Administrator believes may have
necessary information, Congress intended that EPA could gather information from a person not
otherwise subject to CAA requirements. Thus, any limitations that may apply to EPAs
jurisdiction to restrict or otherwise control emissions of air pollutants from sources in the
Western Gulf would not ipso facto also apply to EPAs ability to gather information from those
sources.
Moreover, we do not read the case law as narrowly as commenter. California v. Kleppe, in
which the Ninth Circuit found that OCSLA and the legislative history of 1978 Amendments to
the OCSLA demonstrated that the Secretary of Interior had sole authority to promulgate air
quality regulations for OCS sources, is not dispositive regarding EPAs information gathering
authority. 94
When the Kleppe case was litigated, the CAA did not give EPA any express authority over OCS.
In that pre-1990 case questioning EPAs determination that construction of a floating offshore
storage and treatment facility off the California coast was subject to NSR/PSD requirements and
permitting, the Kleppe court reasoned that Congress had not made its intent clear with respect to
EPA authority over OCS. There, it held that the 1978 Amendments to the OCSLA granted the
Secretary of the Interior the authority to promulgate air quality regulations for the OCS which
94
The Kleppe decision clearly focuses solely on authority to regulate. Throughout, the Court addresses regulations
for the control of air quality. (See, e.g., Kleppe 604 at 1195: [t]he House report [on the 1978 OCSLA
amendments] was silent with respect to EPA jurisdiction over the OCS for control of air quality; [a]t that
time the proposed air quality provisions were amended to specify that the Secretary would promulgate
regulations; the Secretary was to solicit and give due consideration to the views of the Administrator of the
Environmental Protection Agency in developing his regulations; 1196: the statute itself also placed no
apparent limits on the Secretarys regulatory authority over OCS air quality control. (Emphasis supplied.)
943
authority was not to be shared. Kleppe, 604 at 1187 (emphasis added). However, the decision
says nothing about EPAs information gathering authority under CAA Section 114.
By contrast, Congressional intent regarding EPAs information collecting authority has been
quite clear. Congress has consistently broadened EPAs authority to gather information,
extending it to cover even persons not otherwise subject to the CAA. See, Vol. 9 Response to
Comments, Legal Issues on the 2009 final MRR. (Congress repeatedly broadened the scope of
persons subject to 114 first adding any person subject to the CAA in 1977, and then expanding
it to include manufactures of control equipment and any person who the Administrator believes
may have information necessary for the purposes of section 114. See H.R. Conf. Rep. 95-564,
Aug. 3, 1977; P.L. 95-95, 109(d)(3), 113, 305(d) (Aug. 7, 1977); P.L. 95-190, 14(a)(22),
(23) (Nov. 16, 1977); P.L. 101-549, 302(c), 702(a), (b) (Nov. 15, 1990); see also CEDs Inc.
v. EPA, 745 F.2d 1092, 1097(7th Cir.), cert. denied, 471 U.S. 1015 (1984). Thus, section 114 is
not limited to persons who own or operate an emission sources or are otherwise subject to
regulation under the CAA.)
Moreover, Congress made its intentions clear with respect to EPAs authority over the OCS in
the 1990 Amendments to the CAA when it added section 328, OCS Air Pollution to the CAA.
Paragraph (b) of 328 addresses the Western and Central Gulf and states that, the Secretary shall
consult with the Administrator [of the EPA] to assure coordination of air pollution control
regulation for Outer Continental Shelf emissions and emissions in adjacent onshore areas.
Concurrently with this obligation, the Secretary shall complete within 3 years of enactment of
this section a research study examining the impacts of emissions from Outer Continental Shelf
activities in such areas 95 that fail to meet the [NAAQS] for either ozone or nitrogen dioxide.
Based on the results of this study, the Secretary shall, consult with the Administrator, and
determine if any additional actions are necessary. Prior to the 1990 amendments no such
consultation was required under either the CAA or OCSLA. Instead, OCSLA simply required
DOI to cooperate with EPA. 5 U.S.C. 1334(a).
The clear language of the 1990 Amendments indicates that Congress intended that the Secretary
of Interior and Administrator of the EPA each have a role in the Western and Central Gulf.
Indeed, commenter admits that EPA has at least a consultation role regarding sources in the
Western Gulf, but then fails to address whether in fulfilling its role EPA could collect necessary
information. Rather, it merely asserts that EPA has no authority at all to gather information from
sources in the Western Gulf, regardless of whether related to this consultation role,
demonstrating the flaws in their argument. Any limitation of the Agencys express authority to
issue air quality regulations does not mean that it has no authority there.
Further, as pointed out by other commenters, [see, e.g., EPA-HQ-OAR-2009-0923, 1155-13 and
authorities cited therein] it would be unreasonable to interpret OCSLA to somehow implicitly
repeal EPAs information gathering authority under Section 114 as it pertains to offshore
activities. OCSLA contains no such language, while on the other hand, Congress was quite clear
regarding EPAs broad information gathering authority. The argument, as they point out, runs
foursquare into [the] presumption against implied repeals, and implied appeals are not favored.
95
[Portions] of the United States Gulf Coast Outer Continental Shelf that are adjacent to the States not covered by
subsection (a) of this section which are Texas, Louisiana, Mississippi, and Alabama.
944
Only where provisions in two statutes are irreconcilable will an implied repeal be countenanced,
and nothing in OCSLA and Section 114 are irreconcilable. As explained elsewhere, EPA has
valid reasons for gathering this information, consistent with carrying out provisions of the CAA,
and none of these interfere with or undermine DOIs authority under OCSLA. Significantly,
even the Kleppe court acknowledged that the conference committee report on the 1978 OCSLA
amendments stated that it did not intend to affect whatever present authority the Environmental
Protection Agency has in applying and enforcing the Clean Air Act. Kleppe 604 at 1196. 96
Moreover, aside from the question of whether EPA can directly regulate emissions from sources
in the Western and Central Gulf, EPA is not precluded from performing other tasks, like
requesting air emissions information from such sources. This information gathering exercise is
completely different than that struck down by the court in Williams Companies v. Federal
Energy Regulatory Commission, 345 F. 3d 910 (D.C. Cir. 2003). There, FERC was trying to
rely on the open-access provisions of OCLSA for its authority to gather information, and the
court held that open-access provisions did not confer authority upon FERC to issue regulations.
Here, EPA is not relying on OCSLA for its authority, but rather on the CAA, which, as discussed
above and in Vol. 9, above, which unambiguously grants EPA broad authority to gather
information for purposes of carrying out the CAA. 97
Moreover, the information EPA is gathering under this rule from petroleum and natural gas
systems, including those in the Western and Central Gulf, is consistent with the purposes of
section 114, which include carrying out any provision of the CAA. EPA is comprehensively
considering how to address climate change under the CAA, including both regulatory and
nonregulatory options. The information from these and other offshore platforms will inform our
analyses, including options applicable to emissions of any offshore platforms that EPA is
authorized to regulate under the CAA.
9.2
CONFIDENTIAL BUSINESS INFORMATION (CBI)

Organization: The Petroleum Association of Wyoming
Commenter: John Robitaille
General Comments Protection of Sensitive Business Information
On page 56358 of the Federal Register, EPA states they are, collecting owner and operator
information through the Certificate of Representation (40 CFR 98.4). At this time, EPA is not
proposing to assign unique identifiers to the owners and operators because of the complexity of
96
The relevant information gathering provisions of the CAA were in place in 1977. See, Vol. 9, above.
97
Further, as pointed out above, monitoring and reporting rules are not regulation. See also, Reconsideration of
Interpretation of Regulations that Determine Pollutants Covered by Clean Air Act Permitting Programs; Final
Rule, 75 FR 17004 (April 2, 2010)
945
ownership structures (including percentage shares of owners, subsidiaries, holding companies,

and limited liability partnerships) that can be used in the multiplicity of industrial sectors
required to report emission data under this rule.
Many PAW members are privately held companies, and consider percentage of shares of owners
confidential business information.
Response: Todays final rule does not address whether data reported under subpart W will be
released to the public or will be treated as confidential business information (CBI). Please see
Section II.B of the preamble to todays final rule for further information regarding the treatment
of CBI and the Proposed Confidentiality Determination for the Mandatory Greenhouse Gas
Reporting Rule and Proposed Rule Amendment Specifying Procedures for Handling Part 98
Data (EPA-HQ-OAR-2009-0924-001) along with the Proposed Confidentiality Determinations
for Data Required Under the Mandatory Greenhouse Gas Reporting Rule Supplemental
Proposal (EPA-HQ-OAR-2009-0924-008) which are located at
http://www.epa.gov/climatechange/emissions/CBI.html#proposal.
.

Organization: Interstate Natural Gas Association of America
Commenter: Lisa Beal
Optical Gas Imaging Instrument Video Records are Business Confidential and Should Not Be
Available via a FOIA Request
For the optical imaging leak screening, the preamble indicates that video records are required
and 98.237(b) specifies that the operator must retain, Results of all emissions detected and
measurements. EPA should clarify in the rule text whether video records are required for
compliance with 98.237(b).
INGAA understands that EPA still plans to propose amendments to Subpart A to address
Confidential Business Information (CBI). If that action is completed on a timely basis, it should
identify leak survey video records as CBI. If the Subpart A amendments to address CBI are not
finalized prior to Subpart W promulgation, this issue should be addressed in the Subpart W Final
Rule.
EPA discusses the CBI rulemaking in the October 2009 Final Rule for GHG Mandatory
Reporting. At 74 FR 56287, EPA indicates,
Through a notice and comment process, we will establish those data elements that are
emissions data and therefore will not be afforded the protections of CBI. As part of that
exercise, in response to requests provided in comments, we may identify classes of information
946
that are not emissions data, and are CBI.

Due to security concerns associated with national industrial operations and energy infrastructure,
it is imperative that facility video records from optical camera leak surveys are afforded
protection as CBI. Optical camera video records should be considered CBI and should not be
available through a Freedom of Information Act (FOIA) request.
Response: Todays final rule has been revised, EPA does not require retention of a video
recording of the leak detection using optical imaging cameras. Hence, CBI is not relevant in this
case.

Organization: Yates Petroleum Corporation
Commenter:
General Comments Protection of Sensitive Business Information
On page 56358 of the Federal Register, EPA states they are, collecting owner and operator
information through the Certificate of Representation (40 CFR 98.4). At this time, EPA is not
proposing to assign unique identifiers to the owners and operators because of the complexity of
ownership structures (including percentage shares of owners, subsidiaries, holding companies,
and limited liability partnerships) that can be used in the multiplicity of industrial sectors
required to report emission data under this rule.
Yates is a privately held company, and considers percentage of shares of owners confidential
business information.
Response: Regarding the protection of sensitive business information, please see the response to
EPA-HQ-OAR-2009-0923-1015-27.
9.3
RELATIONSHIP TO OTHER CLEAN AIR ACT (CAA) PROGRAMS

Organization: DFW Regional Concerned Citizens (DFWRCC),
Commenter: Faith Chatham
Commenter Type: Industry - oil and gas Without an actual measure to determine the actual
emissions and testing to determine the composition of the emissions per facility (apparatus or
site), there is no way to effectively enforce these rules. Currently each operator can say that
elevated readings are just part of their normal allowed (or calculated (estimated or guesstimated)
25 tons per year of VOCs.
947
I live in Arlington, Texas. Since 2006 `the Arlington City Council and Texas Railroad
Commission has permitted 187 gas wells with accompanying pipelines, storage tanks, and
compression stations. The City of Arlington has 39 completed natural gas drilling applications.
Numerous others have already been heard by the P&Z Commission but have not been heard by
the City Council. Two of these applications currently being considered by the City Council are
for drill sites and Natural Gas pipelines at two sites which are each within three blocks of the
Cowboy Stadium which draws up to 100,000 fans. One is between the stadium and the only
ER/Trauma center in this part of town. If an accident occurs at a well permitted at that site when
there are events at the stadium, 350,000 people could easily be harmed. My estimates include
pupils at a school for the deaf one block from the proposed drill site, patients at a kidney diolisis
center or in a medical office building on two sides of the site, shoppers at the Lincoln Square
Shopping Center in the next block from the site, people driving on I-30 to the immediate north of
the site, residents in a nursing home one block west of the site, and residents in apartments and
homes around the site. When the stadium was built an increase in automobile traffic is believed
to have impacted air quality yet no baseline testing has been done to measure air quality when
there are no events at the stadium and none has been done when the stadium has capacity crowds
to compare to see if there is a significant difference. Without combining the emissions already in
the neighborhood with projected emissions from a gas drilling/production/storage/compression
or transfer facility, it is impossible to protect the health of those near those sites. Cities should be
limited in permitting new facilities when they are out of air quality attainment levels. Vapor
Recovery Systems should be mandated on all O&G tanks in areas within 3000 feet of doctor's
offices, medical clinics, hospitals, parks, schools, day cares, homes and churches. Citizens who
own or rent homes or apartments or lease business space should be notified if that space is within
3000 feet of a proposed O&G drill, storage, transfer or compression site. Notification should be
for a wider footprint than the mere set-back requirements from adjacent building. These
requirements should be enacted for people in every state. States and local municipalities should
be restricted from issuing new O&G permits in areas where the emissions from already existing
facilities are 1. calculated not accurately measured 2 exceed a cumulate level measured in tons
per year per community set by the EPA utilizing sound science which protects the health of
unborn babies, young children, pregnant women and medically vulnerable people. Oil and Gas
drilling should be restricted adjacent to roads and or intersections with traffic counts exceeding a
level set by the EPA based on sound science showing impairment to the common good by a
potential accident at that site.
Oil and Gas permits should have to demonstrate that an accident at that site impacting the
geographic area common by accidents at similar sites will not result in loss of life to thousands
of people within the mile or half mile (as set by EPA based on sound science) footprint of the
well or facility. When permits are issued at high density population sites, the city and operator
should be required to submit a disaster plan including evacuation plans for venues attracting
large numbers of people, for medical facilities and schools and residences for the handicapped or
elderly.
The route of pipeline serving a O&G well site should be disclosed prior to permitting of the well
to residents and business persons living or renting space within 3000 feet of the drill site,
pipeline, storage or compression station. Renters and business owners and patrons of businesses
should be considered when permits are considered instead of just property owners who sign
948
mineral leases.
There should be continuous monitoring of air quality by O&G sites within 1500 feet from day
care centers, hospitals, nursing homes, medical office buildings, shopping centers, schools,
churches, parks, playgrounds, stadiums and residence and along natural gas pipelines with
unodorized gas running in residential neighborhoods and by business or educational
establishments where people regularly.
In Arlington if every wellhead permitted since 2006 is allowed 25 tons of VOCs per well head
and "qualifying apparatus" just the 187 new wellheads already permitted allows them to emit
nearly 5000 tons of VOCs a year. That is without consideration of storage tanks and compression
stations and emissions from pipelines! With addition of venues which attract thousands of cars
into the region and growth in residential and visitor traffic to the city, we cannot continue to
allow 25 tons per apparatus or site of VOCs. There is no requirement that their calculations are
actually measured. There is no way to effectively enforce the proposed rule unless you mandate
devices to measure the VOCs. Companies might be allowed to utilized actual measurements on
sites until there is a demonstration that the amount of their VOCs and the content of the VOCs is
not harming the people's health or contributing to greenhouse gas in levels which when
combined with emissions industries and practices in the area do not create a cumulative negative
effect on human health and the environment.
Water samples, soil samples, and air samples and blood samples from a sample of residents in
the area should be taken before permits are issued and compared periodically to provide data so
that future decisions can be made regarding the real impact of such activities on human health.
Similar studies should be made on flora and wildlife in the area. It would not be necessary to do
this at every well site but it should be mandated in communities which have sharp escalation in
the number of sites permitted in a few years such as Arlington, Fort Worth and other places in
the Barnett Shale where they permit over 50 new sites a year.
Response: EPA has reviewed this comment, and concludes it is out of scope with todays final
rule and The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98).
Commenter: Burckhalter
EPA should not change the definition of "facility" in order to aggregate crude oil and natural gas
production sites.
EPA states that most small businesses will fall under the 25,000 mtCO2e threshold and would not
be required to report GHG emissions; however, by requiring an operator to report emissions
from all of their crude oil and natural gas production sites within a basin as a "facility" will
subject many small businesses to the proposed GHG data collection and reporting requirements
which is directly opposite of EPA's stated goal. The proposed aggregation of crude oil and
natural gas production sites is contrary to the Clean Air Act definition of a "facility" and to
949
subsequent interpretations that industry and regulators understand. Additionally, EPA is not
applying a new definition of facility and aggregating sites from other industry sectors in the same
fashion (with the exception of Subpart RR). By aggregating sites, the impacts to small crude oil
and natural gas businesses will be significant.
Response: EPA does not agree with the commenter. The commenter does not provide any
details on its claim that the rule will subject many small businesses to the proposed GHG data
collection and reporting requirements. EPA through its analysis has in fact determined that the
impact on small businesses will be insignificant. Please see response to EPA-HQ-OAR-2009
0923-1005-7 and Section 5.2 of the EIA for further details.
EPA also does not agree that the onshore facility definition is contrary to the CAA requirements.
Please see Topic 1: GHG Reporting under Subpart W and the Consolidated Appropriations Act
in Volume 9 of the Response to Comments and the response to EPA-HQ-OAR-2009-0923-1044
1 for further details.

However, the State disagrees with the basis for aggregating oil and gas facilities for greenhouse
gas reporting largely because it creates additional inconsistencies between the MRR and
Prevention of Significant Deterioration (PSD) and Title V Operating Permit (Title V) Programs
under the Clean Air Act beyond the existing discrepancies between the rules (i.e. actual vs.
potential to emit, metric vs. short tons, etc.). Specifically, it argues that the rule departs from the
traditional source definition under PSD and Title V and places more emphasis on control of
equipment rather than operation or ownership, that it addresses emissions outside of the PSD and
Title V permitting programs purview and that the facility definition based upon a hydrocarbon
basin disregards local, state or national boundaries.
Response: As we pointed out in Volume 9, Legal Issues, of the EPAs Response to Public
Comments for the Final MRR, to which the commenter is referred, there are myriad reasons
supporting EPAs authority and need to gather information under the rule. In the Advance
Notice of Proposed Rulemaking Regulating Greenhouse Gas Emissions under the Clean Air Act,
74 Fed. Reg. 44354 (July 30, 2008) (ANPR) we articulated the various CAA provisions under
which information gathered about GHGs would be relevant and useful. It is reasonable for the
Agency to collect GHG emissions information from petroleum and natural gas systems to inform
the Agencys regulation and understanding of GHG emissions under any number of those
programs as well as to carry out its Congressional mandate. The PSD and Title V programs are
only two such relevant programs and it is not appropriate to compare data collection under the
MRR to those programs. The MRR is unique, and its purpose is not limited to either of those
programs. That said, information gathered through this rulemaking will be valuable in
determining how best to administer the PSD and Title V programs, but they are not the only
950
programs for which information collected under the MRR might be used. As we have pointed
out previously, this includes the National Ambient Air Quality Standard and State
Implementation Plan programs under 1-7-110 of the Act, New Source Performance Standards
(NSPS) under 111, National Emission Standards for Hazardous Air Pollutants (NESHAP or
MACT) under section 112. Additionally, as has been pointed out elsewhere by other
commenters, the structure of the CAA supports EPAs position. Its authority under Section 114
of the CAA as the basis for the proposed GHG reporting rule is completely independent from
EPAs authority for other CAA programs, for example, from EPAs CAA NSR authority,
including CAA Sections 160-169 and 171-193. As such, EPA regulations promulgated under
these distinct provisions in Section 114 cannot and should not impact the other CAA programs
and vice versa. Therefore, EPAs definition of facility for purposes of the proposed Subpart W
in no way impacts the facility definition for similar sources under the NSR program or other
existing CAA programs.
Additionally, information collected from the petroleum and natural gas systems sector is relevant
to the Agencys understanding of GHG emissions, particularly inasmuch as the sector, by
industry, generates the second largest amount of human-made greenhouse gas emissions in the
United States. Further, as explained in the preamble to the April 2010 proposed rule found in
docket (EPA-HQ-OAR-2009-0923-0002) and accompanying Technical Support Document
(TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923), because of the unique
nature of petroleum and natural gas production facilities, it is reasonable and appropriate in this
rule to define the facility based on the hydrogeologic basin and equipment owned or operated by
the entity holding permits in that basin. The physical boundaries of a production or distribution
owner or operators holdings are not discrete such as for other industry segments. And while the
PSD and NSR programs might apply to stationary sources as defined by regulation, the scope of
CAA section 114 is not so limited. The scope of the rule and the authority on which is based is
not limited to regulated sources and to graft a source definition from other regulatory programs
on the rule is neither appropriate nor reasonable, particularly inasmuch as the rule is intended to
apply to or inform a plethora of programs or purposes. Further, while the rule might place
emphasis on equipment owned or operated within the basin, we pointed out in the preamble to
the proposed rule that we had limited the scope of the rule to the most significant emissions
sources in the sector. Although some of those sources might be owned by other than the owner
or operator, they are ultimately under their control and are their ultimate responsibility to control.
Finally, in todays final rule, we have revised the definition of natural gas processing facilities to
not include gathering lines and boosting stations as an emissions source which should eliminate
some of the commenters concerns respecting aggregation.
These additional inconsistencies between the MRR and PSD and Title V Permit Programs are
sure to create confusion within the regulated community and make the inventory collected
through the MRR inconsistent with permits and permitting inventories. While this confusion may
951
lead to what would otherwise be considered as recordkeeping and reporting noncompliance, once
EPA's proposed Tailoring Rule takes effect, these violations may trigger further action under
EPA's High Priority Violator Guidance.
Response: See response to comment EPA-HQ-OAR 2009-0923-1044-1. Further, as stated in the
preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), the
preamble to todays final rule and in Volume 9, Legal Issues, of the EPAs Response to Public
Comments for the Final MRR, the purpose of the reporting rule is broadly to collect information
to inform the Agencys understanding of greenhouse gases. The rule does not regulate
greenhouse gas emissions and it is not intended to conform to any particular existing or
contemplated regulatory air program, instead its purpose is simply to gather information, among
other things, about the location and magnitude of emissions of greenhouse gases. While this
information might be used in the future to regulate GHGs by permit or otherwise, such is not the
exclusive purpose in collection of this data. And while there might be potential for some
minimal confusion, all programs are not required to be completely consistent. EPA does not
agree that the MRR will be inconsistent with existing permits and permitting inventories,
because the existing inventories deal with non-GHGs whereas the MRR deals with GHGs in
specific. Further, as to the relationship between the rule and any enforcement initiative, EPA
cannot prognosticate or speculate how information collected under the rule might be used in any
discrete or particular enforcement or compliance matter or initiative in the future. Please see
Section VI of the preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40
CFR part 98), and Volume 8, Compliance and Enforcement, of the EPAs Response to Public
Comments, dated September 2009.
See preambles and Volume 9, Legal Issues, of the EPAs Response to Public Comments for the
Final MRR for a more detailed discussion.

Organization: Resolute Energy Corporation
Commenter: Patrick E. Flynn
Annual Report Due Date
Resolute Comments:
Given the vast amount of information that must be collected and compiled, Resolute does not
believe that three months is sufficient time to collect, calculate and QA/QC the data needed to
develop and submit a report. In addition, many industries are already obligated to submit several
data-intensive reports to various agencies, including the EPA, in the first quarter of the year.
These include Title V semi-annual monitoring reports and annual certifications under the CAA;
quarterly deviation reports under the CAA; Discharge Monitoring Reports under the Clean Water
Act; and Tier II reports under the Emergency Planning and Community Right-to-Know Act.
Resolute requests that the EPA consider requiring annual reports to be submitted by June 30th of
each year. This date would be consistent with other registry programs, such as that established by
952
The Climate Registry.

Response: EPA does not agree with the comment. Please see the preamble (Section II.J) of The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), for the response on the
selection of the reporting deadline.
Organization: Enterprise Products
Commenter: Rodney Sartor
Aggregation -The aggregation of gas gathering and processing operations, as proposed in
Subpart W, is inconsistent with current Clean Air Act (CAA) regulations and other subparts of
the greenhouse gas reporting program. It is also incompatible with actual ownership and
operation scenarios for affected facilities.
Response: EPA does not agree with the commenter. However, EPA has not included reporting
of emissions from the gathering and boosting systems in todays final rule. See Section II.E and
II.F of the preamble to todays final rule for further details. For further details on legal authority
to aggregate sources, please see Topic 2: Aggregation of Gathering and Boosting Systems with
Processing Facilities in this document, Volume 9, of the Response to Comments.
Organization: Independent Petroleum Association of Mountain States
Commenter: Kathleen M. Sgamma
In the alternative, if EPA decides to proceed with the definition of Onshore petroleum and
natural gas production facility as proposed, IPAMS requests that EPA expressly state that the
facility definitions proposed in this rule do not impact requirements under other EPA regulations,
for example New Source Review (NSR) in the rule itself, as it did on page 4 of the March 2010
FAQ document for Proposed Subpart W, or at a minimum provide a similar statement in the
preamble to the final rule when it is published in the Federal Register.
Response: With regards to impact of todays final rule on other programs, please see the
response to EPA-HQ-OAR-2009-0923- 1174-5.
953
9.4
COMMENTS RELATED TO PROCEDURES

Organization: Offshore Operators Committee
Commenter: Allen Verret
98.3(c)(11): Legal name(s) and physical address(es) of the highest-level United States parent
company(s) and the percentage of ownership interest for each listed parent company Also, on
page 18613 of the Federal Register, under C. Definition of the Source Category, EPA states that
identifying the facility is clear since there are physical boundaries and ownership structures that
lend themselves to identifying scope of reporting and responsible reporting entities.
OOC Comment: While the facility itself may have a single or multiple owners, each
development and/or production field tied back to the facility may have multiple owners as well.
This may significantly complicate the determination of percentage of ownership interest for each
reporting entity. We feel the EPA should more clearly define the reporting expectations of
ownership specific to offshore petroleum and natural gas facilities, and either remove the
requirement to report percent ownership for offshore facilities, or clearly define facility
ownership as applying only to the facility itself, and not its associated production fields.
Response: EPA has reviewed this comment. Please see Section 98.3 and 98.4 of The Final
Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), which address reporting
requirements and ownership issues. Furthermore, todays final rule amends Section 98.2 on who
must report, requiring reporting from owners and operators of any facility that is located in the
United States or under or attached to the Outer Continental Shelf (as defined in 43 U.S.C.
1331) and contains definitions for offshore and the outer continental shelf.

EPA should require that annual reports be submitted by June 30th of each year.
Response: EPA disagrees with the comment and is retaining the annual reporting due date of
March 31. Please see the preamble (Section II.J) of The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98), for the response on the selection of the reporting deadline.
9.5
COMMENTS ON PORTABLE EQUIPMENT REPORTING REQUIREMENT
954

Organization: Chesapeake Energy Corporation
Commenter: Grover Campbell
Emissions from Portable Non-self-propelled Equipment Should be Exempt Under Subpart W.
EPA should not require reporting emissions from non-stationary and portable sources under
Subpart W for the following reasons.
First, EPA is requesting the collection of this data under its Section 114 authority. Section 114 of
the CAA allows the Administrator to "require any person who owns or operates any emission
source . . . or who the Administrator believes may have information" to provide it in response to
an information request. 42 U.S.C. 7414 (emphasis added). Non-stationary and portable
equipment is typically owned by contract service companies, not site operators. Moreover, in
many cases, site operators do not have the operational data or the necessary records to perform
the calculations necessary to certify the accuracy of GHG emissions data. Regardless of whether
the equipment is stationed at a well site for more than 30 days, this equipment is often moved
from well to well and between operators. Furthermore, because the site operator does not own
this equipment, the site operator does not control its operation, does not perform or schedule its
maintenance, and is not responsible for collecting the fuel consumption data. In short, EPA has
greatly underestimated the difficulties site operators will face trying to track equipment and
estimate emissions from field portable equipment that is owned and operated by third-parties.
Therefore, consistent with its Section 114 authority and other provisions of the CAA, EPA
should adopt provisions that determine applicability based on whether the party owns and/or
operates a source. See, e.g., 42 U.S.C. 7414(a)(1); 42 U.S.C. 7410(a)(2)(F) (requiring owners
and operators of stationary sources to install maintain and replace equipment in order to monitor
emissions from sources); 40 C.F.R. 60 .752 (establishing air emissions standards for owners
and operators of municipal solid waste landfills).
Second, requiring petroleum and natural gas systems to report emissions from portable
equipment under Subpart W would be inconsistent with EPA's approach elsewhere in the MRR.
Requiring site operators to report these emissions under Subpart W places a disproportionate
burden on the petroleum and natural gas systems sector because other provisions in the MRR,
such as Subpart C, provide an express exemption for portable equipment, emergency generators,
and emergency equipment. See 74 Fed. Reg. at 56,289 (Oct. 30, 2009). Similarly, the MRR does
not require the reporting of emissions from other types of construction activities, such as
housing, commercial building, or roadways. See 74 Fed. Reg. at 56,266-67 (Oct. 30, 2009).
Third, EPA's Section 114 authority is limited to emissions from stationary sources. 42 U.S.C .
7414(a). The term "stationary source" is defined as "any building, structure facility, or
installation which emits or may emit any air pollutant." 42 U.S .C. 741 1(a)(3). It does not
include portable equipment. Id EPA has cited no authority for extending the definition of
stationary source to portable equipment that is stationed at a wellhead for more than 30 days in a
reporting year. See 75 Fed. Reg. at 18,637 . Even if EPA were to adopt this provision despite it
being contrary to law, the practical implication of imposing a 30 day limit on the equipment is
that some site operators would change out the pieces of portable equipment onsite every 29 days
to avoid these recordkeeping and reporting obligations.
955
Fourth, to require petroleum and natural gas systems to report emissions from portable
equipment would also result in double counting of emissions and inaccuracies in the inventory.
Providing an exemption for this equipment under Subpart W is reasonable because the vast
majority of these engines are fired by diesel fuel and their emissions will be reported by suppliers
of petroleum products under Subpart MM. Fifth, requiring estimates of the emissions from
portable and leased equipment would greatly increase the reporting burden on the oil and gas
industry without a corresponding increase in emissions reporting coverage. Emissions from
portable equipment are such a small portion of the overall inventory that "the reporting of [these]
GHG emissions is unreasonable given the cost of monitoring and the relative level of GHG
emissions." See 74 Fed. Reg. at 56,291 (Oct. 30, 2009).
Finally, EPA could use other alternatives to estimate the emissions from this equipment because
the total number and type of wells completed in the U.S. are well known facts. For instance, EPA
could easily obtain a reasonable estimate of those GHG emissions using the Natural Gas Star
data. The data provided from the Natural Gas Star program could be added to EPA's inventory to
get a more complete emissions estimate for drilling and completing wells in the inventory.
Therefore, it is unnecessary for EPA to impose this additional and disproportionate burden on
petroleum and natural gas systems by requiring facilities to report emissions from portable
equipment.
Response: First, EPA disagrees with the commenter on its authority to collect information from
contracted equipment. Please see the response to EPA-HQ-OAR-2009-0923-1031-21 for further
details.
Second, the EPA does not agree that requiring emissions from portable equipment is inconsistent
with the general approach in the MRR. For further details, please see the response to EPA-HQ
OAR-2009-0923-1015-35.
Third, In todays final rule, EPA has removed the 30-day at wellhead clause for all portable
equipment that have to be monitored. For further details, please see the response to EPA-HQ
OAR-2009-0923-1170-7. EPAs authority under Section 114 is not limited to collection of
information from stationary sources as defined by the PSD/NSR program. See Volume 9,
Legal Issues, of the EPAs Response to Public Comments for the Final MRR.
Fourth, EPA is not excluding combustion emission under Subpart W because of double counting
under Subpart MM. For further information, please see the response to EPA-HQ-OAR-2009
0923-1042-26.
Fifth, EPA has clarified language and provided equipment threshold for external combustion
equipment that will result in a reasonable burden on reporter to monitor portable, contracted,
rented, and leased equipment. Please see response to EPA-HQ-OAR-2009-0923-1151-104 for
further details.
Finally, EPA cannot use data from the Natural Gas STAR Program. For further information
please see the response to EPA-HQ-OAR-2009-0923-1004-2.
956

Organization: Western Climate Initiative (WCI)
Commenter: Michael Gibbs
The WCI supports the proposed approach in SECTION 98.230(a)(2) of including portable non
self-propelled equipment within scope of onshore petroleum and natural gas production in
Subpart W. Portable equipment may account for a substantial portion of emissions in this
industry segment. However, WCI recommends that all portable equipment be included, rather
than only the equipment stationed at a wellhead for more than thirty days, as proposed in
SECTION 98.231(b). Portable equipment used in drilling, completions and workovers may often
be on site for less than thirty days, but nevertheless collectively account for significant
emissions. Because onshore production entities will report emissions at the basin level,
recordkeeping would be simplified and verification would be easier without the necessity for
tracking the time each piece of portable equipment was at a given wellhead. Given that the
impacts of greenhouse gases depend primarily on global rather than local concentrations, the
usual rationale for ignoring short-term temporary sources does not apply here.
Response: Please see the response to EPA-HQ-OAR-2009-0923-1040-5 for more details on
reporting requirements for portable equipment.
EPA has removed the 30-day at wellhead clause for all portable emissions sources. For further
details, please see the response to EPA-HQ-OAR-2009-0923-1170-7.

Organization: Anadarko Petroleum Corporation
Commenter: William W. (Bill) Grygar
The vast majority of engines included in this category are fired by diesel fuel. The amount of
GHG emissions (both point and fugitive) from these portable non-self propelled equipment
sources would be from combustion of diesel fuel. Emissions from non-road engines are
exempt from reporting under Subpart C. Reporting GHG emissions from these engines
under Subpart W would result in double counting of emissions since emissions from the
combustion of diesel fuel are reported by suppliers of petroleum products under Subpart
MM.
Response: EPA would like to clarify that since the emissions from portable combustion sources
in onshore production is large, Subpart W requires the reporting of onshore production portable
combustion emissions under Subpart W in todays final rule, overriding the exemption in
Subpart C. EPA is not excluding combustion emission from engines under Subpart W because of
double counting under Subpart MM. For further information, please see the response to EPA
HQ-OAR-2009-0923-1042-26.
957

References to portable equipment: It is apparent that EPA expects emissions from portable
equipment at production wellheads to be reported, but not portable equipment from other
segments or applications. 98.230(a)(2) includes portable equipment for onshore production and
98.231(b) indicates emissions should be reported for portable equipment stationed at a
wellhead. For clarity and to avoid confusion with underground natural gas storage wells, INGAA
recommends revising 98.231(b) indicate, You must include combustion emissions from
portable equipment that cannot move on roadways under its own power and drive train and that
is stationed at an onshore production wellhead
Response: EPA agrees with the commenter. EPA has clarified that the reporting of portable
equipment applies only to onshore production. This clarification has been provided in Section
98.232 of the rule.
Organization: Colorado Oil and Gas Association
Commenter: Tisha Conoly Schuller:
98.231 Reporting threshold.
Including emissions from drilling rigs located on a well for more than 30 days is unnecessary.
See paragraph one under comments listed under 98.230 above.
Response: EPA disagrees with the commenter on emissions from drilling rigs. Please see the
response to EPA-HQ-OAR-2009-0923-1170-7 and EPA-HQ-OAR-2009-0923-1015-35 for
further details.
Commenter:
General Comments Portable and Standby Equipment
EPA states, For applying the threshold defined in 98.2(a)(2), you must include combustion
emissions from portable equipment that cannot move on roadways under its own power and
drive train and that is stationed at a wellhead for more than 30 days in a reporting year, including
drilling rigs, dehydrators, compressors, electrical generators, steam boilers, and heaters. This
requirement is unclear. Must it be the same piece of equipment (i.e., what if there are three
different portable engines that together add up to >30 days)? It is important to note that
equipment often moves from well site to well site, however, as written in the rule it appears that
equipment must be tracked by basin. While it is common to track equipment by asset or business
line, crossing basins will make this reporting requirement difficult to comply as it is information
not currently tracked in this manner.
958
98.234(a) will require standby equipment to be included in annual leak detection which is overly
burdensome: by nature, standby equipment is operated in short intervals, and is often operated in
place of other equipment that would otherwise be emitting GHG emissions. Also, this
requirement would force YPC field personnel to make multiple trips to each wellhead each year
to quantify emissions that are otherwise minimal.
Response: EPA has removed the 30-day at wellhead clause for all portable emissions. For
further details, please see the response to EPA-HQ-OAR-2009-0923-1170-7. In regards to
tracking equipment by basin, please see response to EPA-HQ-OAR-2009-0923-1060-4.
In todays final rule, the use of word standby is used only to refer to an operational mode for a
compressor. EPA has determined that most reporters understand the term standby with regard to
compressors. EPA has clarified the term in todays final rule. Please see the response to EPA
HQ-OAR-2009-0923-1206-59. EPA has also clarified in todays final rule that onshore
production compressors only need to use an emissions factor to estimate process emissions
(vented and equipment leaks) no measurements are needed in the three modes of operation.
Section 98.231(a) Reporting threshold. Section 98.231(a) says to report emissions from
petroleum and natural gas systems defined in Section 98.230. Section 98.231(b) says in applying
the reporting threshold, include emissions from portable equipment that are stationed at a well
head for more than 30 days in a reporting year. Section 98.230 does not include the same 30 day
provision. Rather Section 98.230(a)(1) for onshore facilities says the facility includes portable
non-self-propelled equipment (including by not limited to well drilling and completion
equipment, workover equipment, gravity separation equipment, auxiliary non-transportation
related equipment, and leased, rented or contracted equipment). EPA should clarify that
emissions associated with portable equipment stationed at a well-head less than 30 days are not
reported. Reporting of portable equipment stationed on a well head less than 30 days will
significantly add to the already burdensome requirements of the rule.
Response: EPA has removed the 30-day at wellhead clause for all portable emissions sources.
For further details, please see the response to EPA-HQ-OAR-2009-0923-1170-7. EPA has
clarified language and provided equipment threshold for external combustion equipment that will
result in a reasonable burden on reporter to monitor portable, contracted, rented, and leased
equipment. Please see response to EPA-HQ-OAR-2009-0923-1151-104 for further details.

Organization: Devon Energy Corporation
Commenter: Richard Luedecke
959
Portable Equipment Combustion Emissions

Devon contends that tracking and estimating emissions from portable combustion devices (e.g.,
drilling rigs, dehydrators, compressors, electrical generators, steam boilers, and heaters) that are
stationed at a wellhead for more than 30 days in a reporting year would be resource intensive and
complicated since this equipment is moved from well to well and between operators several
times during the year. In many cases, the equipment is operated by contractor service companies
and the facility owner or operator would, in most cases, not have the necessary records to
estimate emissions and to certify the accuracy of GHG emissions data from these portable
sources.
In order to minimize the burden of reporting emissions from third party owned/operated sources,
we suggest limiting this requirement to the most prevalent sources, specifically drilling rigs and
rental compression.
Response: EPA has removed the 30-day at wellhead clause for all portable emissions sources.
For further details, please see the response to EPA-HQ-OAR-2009-0923-1170-7.
In todays final rule, EPA has restricted the list of portable combustion equipment required to
report. However, EPA disagrees with limiting the list to just drill rigs and rental compression.
For further details, please see Section II.F of the preamble. To further decrease the burden on the
industry, EPA has provided an equipment threshold for external combustion equipment in
onshore production and natural gas distribution. For further clarification, please see the response
to EPA-HQ-OAR-2009-0923-1024-23. EPA has clarified language and provided equipment
threshold for external combustion equipment that will result in a reasonable burden on reporter to
monitor portable, contracted, rented, and leased equipment. Please see response to EPA-HQ
OAR-2009-0923-1151-104 for further details.
960
9.6
COMMENTS ON LEASED/RENTED EQUIPMENT REPORTING

REQUIREMENT

Organization: Natural Gas Supply Association
Commenter: Patricia W. Jagtiani
Because Reporting Responsibility for Onshore Natural Gas Production Can Be Complicated,
EPA Should Provide Safeguards
The proposed Subpart W overlooks significant complications associated with determining
reporting responsibility for onshore production wells. A single well can be owned by one entity,
be operated by another entity, lease portable equipment from a third entity, and have that
portable equipment operated by yet another entity. In these situations, the entities that directly
operate certain equipment are in the best position to gather emissions data for that equipment,
whereas other entities working at the same well site have limited ability to verify that data. Yet
the proposed rule places the burden of reporting entirely on the owner of the well or the holder of
the operating permit. This requirement could place onshore natural gas production entities in the
impossible situation of being held to account for errors or omissions committed by third parties.
NGSA respectfully requests that EPA address this concern by taking the following steps:
a. Allow Reporting Entities to Reasonably Rely on Operators. NGSA is
concerned that in cases where a reporting entity must collect emissions data from other
entities operating equipment at a particular well, the reporting entity could be held liable
for non-obvious errors or omissions committed by those other entities. The Reporting
Rule should allow reporters to reasonably rely on data supplied by operating companies
associated with the production well. This safe harbor would not, of course, apply
where reliance is unreasonable, such as when the reporting entity has reason to know that
data is erroneous or false.
b. Disallow Vicarious Liability for Errors Committed by Reporting Entity.
NGSA is also concerned that entities in the onshore natural gas production sector that
provide emissions data to reporting entities may then be held liable for subsequent errors
or omissions committed by that reporting entity. EPA should clarify that an entity that
provides properly collected emissions data to a reporting entity will not be held liable if
that reporting entity commits an error or omission in reporting that data to EPA.
c. Extend annual report submission deadline for onshore production facilities.
NGSA is concerned that the annual deadline of March 31 for submission of emission
reports will not allow sufficient time for production well owners and operators to collect
and compile all necessary data from all entities responsible for operating the well.
Therefore, NGSA respectfully requests that EPA extend the annual reporting deadline by
90 days for entities in the onshore production sector.
961
Response: Regarding item a, allow reporting entities to reasonably rely on operators, please see
Section II.F of the preamble in todays final rule for a discussion of the role of the Designated
Representative and also comment response EPA-HQ-OAR-2009-0923-1167-39.
Regarding vicarious liability, the focus of the rule is on reporting by owners or operators through
a Designated Representative or Alternate Designated Representative and not third parties who
might provide emissions data to the reporting entity. Owners or operators are free to provide or
arrange for any third party indemnification which they might feel appropriate. See also, Section
VI of the preamble to the October 2009 Greenhouse Gas Reporting Rule.
Concerning extension of the reporting period, EPA has determined not to extend the deadline.
However, please also refer to Section II.F of the preamble in todays final rule for a discussion of
data reporting requirements and the option to use Best Available Monitoring Methods.
The designated representative (DR) is the entity that is responsible for submitting the emissions
data pursuant to todays final Rule. Please see the response to EPA-HQ-OAR-2009-0923-1024
16.
Comment Number: EPA-HQ-OAR-2009-0923-3568.4-5

Commenter: Karen Ritter

Fourth, another key issue for the industry is its ability to track and estimate emissions from field
portable combustion devices that are stationed at a wellhead for more than 30 days in a reporting
year. EPA should require a less resource intensive and complicated approach that recognizes that
this equipment is moved from well to well or between operators several times during the year.
Similarly, API does not agree with EPAs extension of the reporting requirements to include
rental and other equipment that is neither owned nor operated by the reporter.
Response: EPA took all comments on portable equipment into consideration and has removed
the 30-day at wellhead clause in todays final rule to avoid practical issues with determining the
time the portable equipment is at the wellhead. Please see the response to EPA-HQ-OAR-2009
0923-1170-7 for further details.
EPA disagrees with the commenter to exclude rental and other equipment from the reporting
rule. For further details, please see the response to EPA-HQ-OAR-2009-0923-1031-21.
9.7
COMMENTS ON REPORTING OF CONTRACTOR EMISSIONS

Organization: American Exploration & Production Council
Commenter: V. Bruce Thompson
962

3. Emissions from portable non-self-propelled equipment (such as well drilling and completion
equipment, workover equipment, gravity separation equipment, auxiliary non-transportation
related equipment) do not need to be reported.
EPA should delete the reporting of emissions from these non-stationary and portable sources.
EPA should simply use its own updated total emissions estimates for these sources. The huge
added expense for industry to calculate, keep records, and report emissions from thousands of
these temporary sources is not justified by the small incremental emissions reporting accuracy
EPA many be expecting from this approach. More specifically:
Most wells arc drilled and completed using contract service companies. EPA underestimates
the difficulty of tracking equipment and estimating emissions from field portable equipment that
are operated by third parties even when stationed at a well site for more than 30 days. Complying
with this requirement as proposed would be very resource intensive and complex since this
equipment is often moved from well to well and between operators. In many cases, site operators
would not have the operational data to perform the needed calculations and/or the necessary
records to certify the accuracy of GHG emissions data. When the site operator does not control
the operation or maintenance of the equipment, it is not appropriate to require reporting and
compliance tasks of the well site operator because they do not participate in engine maintenance
or the collection of fuel use data. All other Clean Air Act programs establish applicability based
on whether a party owns and/or operates a source because it is not feasible for someone who
does not control the day-to-day operation of a source to collect the required information or
monitor the source's usage. No other industry sector is required to report contractor's emissions
for this reporting rule.
All of these activities are non-stationary source and temporary construction activities that
should be excluded just as other construction activities (housing, commercial building, roadways,
etc) are excluded. Requiring estimates of these emissions greatly increases the reporting burden
on the oil and gas industry without a corresponding increase in emissions reporting coverage.
The total number and type of wells completed in the U.S. is well known; therefore, EPA could
easily obtain a reasonable estimate of those GHG emissions using the Natural Gas STAR data
mentioned above. EPA could then add-on the emissions estimates for drilling and completing
wells to the inventory.
The vast majority of engines included in this category are fired by diesel fuel. The vast amount
of GHG emissions (both point and fugitive) from these portable non-self propelled equipment
sources would be from combustion of diesel fuel. Emissions from non-road engines are exempt
from reporting under Subpart C. Reporting GHG emissions from these engines under Subpart W
would result in double counting of emissions since emissions from the combustion of diesel fuel
are reported by suppliers of petroleum products under Subpart MM.
Response: EPA disagrees with the commenter on emissions from portable equipment. The
emissions contribution from portable equipment is significant enough to warrant data collection.
963
Please see the responses to EPA-HQ-OAR-2009-0923-1015-35 and EPA-HQ-OAR-2009-0923

1170-7 for further details.
EPA has removed the 30-day clause from todays final rule. Please see response to EPA-HQ

Contractor Emissions The WCI supports the EPAs proposed inclusion of contracted equipment
within the onshore petroleum and natural gas production definition. The inclusion of contracted
equipment is critical to ensuring that facilities with different operational structures have equitable
coverage in a reporting (or cap and trade) program and that a complete profile of emissions from
the production sector is obtained. Owners or operators will have the ability to include within
contract terms that contractors must provide sufficient data to enable a complete greenhouse gas
emissions report to be submitted. The WCI also notes that as greenhouse gases are a pollutant,
overall responsibility for their release would seem to rest with the owner of the facility, instead
of with contracted operators.
Response: EPA agrees with the comment. For a more detailed discussion of the inclusion of
contracted equipment, please see Section II.E or Section II.F of the preamble of todays final
rule.
Most wells are drilled and completed using contract service companies. EPA underestimates the
difficulty of tracking equipment and estimating emissions from field portable equipment that are
operated by third parties even when stationed at a well site for more than 30 days. Complying
with this requirement as proposed would be very resource intensive and complex since this
equipment is often moved from well to well and between operators. In many cases, site operators
would not have the operational data to perform the needed calculations and/or the necessary
records to certify the accuracy of GHG emissions data. When the site operator does not control
the operation or maintenance of the equipment, it is not appropriate to require reporting and
compliance tasks of the well site operator because they do not participate in engine maintenance
or the collection of fuel use data. All other Clean Air Act programs establish applicability based
monitor the sources usage. No other industry sector is required to report contractors emissions
for this reporting rule, and the oil and gas industry should not be subject to a different standard.
964
Response: EPA does not agree with the commenter. With regards to tracking and gathering
relevant information on third party equipment and contractor equipment, please see the response
to EPA-HQ-OAR-2009-0923-1170-7 for further details.
As articulated in the preamble to the Final Mandatory GHG Reporting Rule (Final MRR), (40
CFR part 98) as well as in the MRR Response to Comments, Vol. 9, Legal Issues, the scope of
Section 114 is quite broad and applies beyond persons who own or operate an emission source or
who are otherwise subject to regulation under the CAA. Its purpose extends to the collection of
information for carrying out any provision of the Act (emphasis supplied). We note, however,
that subpart W does not require a contractor or third party to report emissions; that responsibility
lies with the owner or operator of the applicable subcategory and industry sector. Section 114
(a)(1) provides that the Administrator may require any person who owns or operates an
emissions source to collect and provide relevant CAA information to EPA. As explained, the
owner or operator is in the best position to collect and report that information to the Agency.
The fact that emissions might be generated by equipment or processes owned by a party
independent of the owner or operator of a particular facility is of not determinitive. The owner
or operator has ultimate responsibility for the facility and the obligation and responsibility to
assure that the required information can be collected through legally binding contracts, leases, or
other agreements. This position that an owner or operator, through contract or other third-party
agreement, cannot otherwise shirk CAA responsibility has long been that of the Agency(see
authorities cited below). In recognition, however, of the fact that such contractual, lease or other
binding provisions might not be in place in all instances, the rule allows a period of time during
which best available monitoring methods may be applied. See Section II.F of the preamble to the
final Subpart W rule for further details.
Although the MRR is not delimited by other CAA programs such as NSR or Title V, and EPA
deems it inappropriate to tailor the provisions of the reporting rules to any specific program, we
have explained this concept numerous times in the context of NSR or Title V which is here
instructive and of precendtial value. The premise underlying this position lies in the notion that
the owner or operator has the power to control, the power of authority to guide, manage, or
regulate the pollutant-emitting activities of those facilities. See, Guidance for Major Source
Determinations at at Military Installations under the Air Toxics, New Source Review, and Title
V Operating Permit Programs of the Clean Air Act in Memorandum for Major Source
Determinations for Military Installations, John S. Seitz, Director, Office of Air Quality Planning
and Standards, August 2, 1996. Therein it is explained that because the contracting entity can
control the contract operators performance through contract terms, contract-for-service activities
are considered part of a source. Likewise, the Agency has flatty stated its policy that temporary
and contractor-operated units [must] be included as part of the source with which they operate or
support and that there are no provisions in title I or title V of the Act, or in regulations
developed pursuant to them, for excluding contracted or temporary operations in defining major
sources. Letter from John S. Seitz to Lisa J. Thorvig, Division Manager, Minnesota Pollution
Control Agency, November 16, 1994. The underlyling rationale for including sources over
which an owner or operator exercises control in making major source determinations in the
permitting context is thoroughly explained in 1995 guidance, Letter from William Spratlin,
965
Director, Air, RCRA, and Toxics Division, EPA Region VII, to State and Local ir Directors,
September 18, 1995
A more detailed discussion of this issue is found in Section II., Petroleum and Natural Gas
Systems, of the preamble.
Section 98.233(j) Onshore production and processing storage tanks. The rule indicates that this
source applies to emissions from atmospheric pressure storage tanks receiving produced liquids
from onshore petroleum and natural gas production facilities (including stationary liquid storage
not owned or operated by the reporter.) Noble believes that compliance with the requirement for
including stationary liquid storage not owned or operated by the reporter is overly burdensome
and will not be practical because a reporter may not have access to the necessary operational data
(i.e. throughput) nor legal access to the tankage to collect required samples (e.g. sales oil for API
gravity and Reid vapor pressure) to perform the required calculations. Therefore, Noble requests
that this section only applies when the tank is owned or operated by the reporting entity.
Response: EPA has revised Section 98.233(j) to make it clear that regardless of ownership of
atmospheric pressure storage tanks receiving produced liquids from onshore petroleum and
natural gas production facilities (including stationary liquid storage), it is the onshore petroleum
and natural gas production owner or operator who must report. For further details on legal
authority to require reporting from leased, rented, and contracted equipment, please see the
response to EPA-HQ-OAR-2009-0923-1031-21.
In addition, EPA provided methods in todays final rule that do not require the operational data
from contractors (other than beneficial use or flaring of tank vapors) to estimate emissions. For
example, the reporter can sample the low pressure separator oil and assume that all of the GHG
dissolved in the oil is emissions. See response to EPA-HQ-OAR-2009-0923-1305-48 for further
details on this option.
This rule will require owners of a site to report GHG emissions resulting from rental and portable
equipment located at a well head. Often, the owner of the site does not have emission
information, or relies on information provided by a rental company (i.e., Compressor Systems,
Inc.) that may or may not be accurate or acceptable by EPA standards. Note that 98.4(e)
requires that emissions reports be certified by a Designated Representative. The Designated
Representative must therefore ultimately certify emissions for units they do not control, and there
is no guarantee those emissions are auditable, verifiable, etc. IPANM requests that the EPA
966
determine that emissions from equipment that is on-site, but not under common ownership, does
not need to be reported by the site owner.
Response: EPA disagrees with the commenter. With regards to the legal aspect of reporting
emissions from equipment that is not under common ownership, please see the response to EPA
HQ-OAR-2009-0923-1031-21. In regard to collecting relevant data and reporting from portable
equipment, see response to EPA-HQ-OAR-2009-0923-1170-7 for further details.

Organization: North Slope Burough
Commenter: Edward S. Itta
Similarly, EPA should include all emissions from contracted equipment in the reporting
requirements for the final ruleparticularly in light of the extensive outsourcing within the
industry. These emissions should be accounted for and aggregated with the contracting operator
for threshold determinations.
Response: EPA agrees with the commenter and has retained the requirement for reporting of
emissions from contracted equipment. Please see Section II.D of the preamble of todays final
rule for more details.

As is common practice in the industry, operators contract with service companies to drill and
complete their wells. Complying with this requirement as proposed would be very resource
intensive and complex since this equipment is often moved from well to well, between operators,
and even between geologic basins. In addition, since this equipment is controlled and maintained
by third-party contractors, operators do not have access to the necessary information or data to
report their emissions, or to certify as to the quality of any data that they might be able to obtain.
All other CAA programs establish applicability based on whether a party owns and/or operates a
source because it is not feasible for someone who does not control the day-to-day operation of a
source to collect the required information or monitor the sources usage. No other industry sector
is required to report contractors emissions under EPAs MRR, and it should not be required
under proposed Subpart W. IPAMS requests that EPA remove this requirement.
Response: EPA disagrees with the commenter. With regards to the legal aspect of reporting
emissions from equipment that is not under common ownership, please see the response to EPA
HQ-OAR-2009-0923-1031-21 As regards collecting relevant data and reporting from portable
equipment, see response to EPA-HQ-OAR-2009-0923-1170-7 for further details.
967

Organization: State of New Mexico
Commenter: Jim Norton
2) Defining contractors as reporting entities would create problems with compliance and
coverage of emissions.
Some commenters may suggest, as an alternative approach, that contractors themselves should
be defined as reporting entities and be required to report their emissions directly to EPA. We
consider this to be very problematic from the perspectives of compliance and coverage of
emissions.
Proposed Subpart W currently defines reporting entities for onshore oil and gas production based
on the holding of state drilling permits or state permits to operate oil and gas wells. This provides
a known universe of potential reporting entities, which aids in ensuring compliance. The drilling
or operating permit determines the accountability for the site. Companies which provide rental,
leased or contracted equipment are not generally required to have permits specific to oil and gas
production, which means that the universe of potential reporters who are contractors is unknown.
This would greatly hinder effective compliance activities by EPA or state agencies.
Compliance would also be hindered by the transient nature of the emitting activities of
contractors, given that some of the activities may occur at a given location for only a short period
of time. Most regulatory and tracking systems are based on a specific physical location, and
determining compliance for an entity which might operate a number of different locations, each
for a short period of time, would be extremely difficult.
Companies providing rented, leased and contracted equipment range widely in size, from large
multinational corporations to numerous small businesses. Many of the smaller companies would
not exceed the 25,000 metric ton CO2e reporting threshold, and these unreported emissions may
be substantial.
Response: EPA has retained the requirement for reporting of contracted equipment emissions in
todays final rule and clarified EPA position on this issue; please see the response to EPA-HQ
OAR-2009-0923-1031-21.

Organization: State of New Mexico
Commenter: Jim Norton
3) Owners and operators can control relevant aspects of the contractor's performance through
terms of the contract.
Some commenters may assert that owners/operators cannot be held responsible for operations
and data acquisition pertaining to rented, leased or contracted equipment. We believe there is
968
ample precedent for owners/operators to obtain needed data from contractors through terms of
the contract.
We note that EPA has addressed policy issues related to contractor emissions in the context of
PSD and Title V programs. In several guidance letters, EPA has consistently stated that
"contractor-operated units, must, be included as part of the source with which they operate or
support" Contract/Temporary Operations and Title V, 11-16-94, in Memoranda for Operating
Permits (Title V) -Policy & Guidance Memos, US EPA (www.epa.gov/ttn/oarpg/t5pgm.html).
Although this policy is not necessarily binding on Subpart W of the Mandatory Reporting Rule,
EPA's rationale for this policy is relevant: "the contracting entity can control the relevant aspects
of the contract operator's performance through terms of the contract (e.g., the level of production,
the requirement to implement and maintain emission control measures, the requirement to
comply with all applicable environmental regulations, etc.)" Guidance for Major Source
Determinations at DOD, 8-2-96, in Memoranda for Operating Permits (Title V) -Policy and
Guidance Memos, US EPA (www.epa.gov/ttnloarpg/t5pgm.html). These EPA policies have been
in place for many years, and sources in these air quality programs have been successfully held
accountable for contractor compliance with environmental regulations.
We also note that holding owners/operators accountable for emissions from rented, leased or
contracted equipment provides the owners/operators with several additional opportunities for
emission reductions.
To the extent that any oil and gas sources are subject to EPA Tailoring Rule for greenhouse
gases, it is highly likely that the EPA policy on contractor emissions at Title V source will apply
in implementation of the Tailoring Rule. For consistency, the same policy should be applied to
the reporting of GHG emissions.
We also note that vented emissions of natural gas are already considered the responsibility of the
owner/operator under oil and gas conservation regulations and in common industry practice.
Response: EPA has retained the requirement for reporting of contracted equipment emissions in
todays final rule and clarified EPA position on this issue; please see the response to EPA-HQ
OAR-2009-0923-1031-21.
9.8
OTHER LEGAL COMMENTS

Organization: Environmental Defense Fund
Please accept these comments on behalf of Environmental Defense Fund and our hundreds of
thousands of members and supporters.
We greatly appreciate the opportunity to comment on EPAs Proposed Mandatory Reporting of
969
Greenhouse Gases for Petroleum and Natural Gas Systems. We support the Agencys proposal to
require emissions reporting from the petroleum and natural gas sector, as required, and have
separately submitted comments addressing the technical aspects of this proposal. Here, we write
to underscore the critical nature of economy-wide greenhouse gas emissions data, including
emissions data from the petroleum and natural gas sector. Emissions data are foundational
serving as the cornerstone for effective air quality planning and management; providing muchneeded transparency for Americans concerned about global warming pollution in their
communities; and making it possible to hold the largest polluters accountable for their
greenhouse gas emissions.
Policymakers and the public have already lost a years worth of this critical data, and EPA must
act swiftly to satisfy its legal duty to require reporting from the petroleum and natural gas sector.
In two successive enactments, Congress pointedly instructed EPA to require mandatory reporting
of greenhouse gas emissions for all sectors of the economy of the United States no later than
June 26, 2009. See Fiscal Year 2008 Consolidated Appropriations Act, Pub. L. No. 110-161, 121
Stat. 1844, 2128 (Dec. 26, 2007) and Appropriations Act of 2009, Pub. L. No. 111-8, 123 Stat.
524, 729 (March 11, 2009). Congress thus called for a comprehensive program that encompassed
not a few, not some, but rather all sectors of the economy. EPAs failure to promulgate final
reporting requirements for petroleum and natural gas systems (40 CFR, Part 98, Subpart W)
constitutes an ongoing failure to perform an act or duty under the Clean Air Act (CAA), 42
U.S.C. 7401, et seq., that is not discretionary with the Administrator within the meaning of
304(a) of the CAA, 42 U.S.C. 7604(a). We therefore urge EPA to take all final steps necessary to
complete this rulemaking in a prompt manner and to fulfill its legal duty.
EPA possesses broad statutory authority to satisfy this legal duty and to require emissions
reporting from the petroleum and natural gas sector. CAA sections 114 and 208 provide EPA
with long-standing and expansive information-collection authority. See 42 U.S.C. 7414, 7542.
Moreover, through successive revisions to its applicability and scope, Congress has broadened
section 114 to include persons not previously covered by [the section]. See H.R. Conf. Rep.
95564, Aug. 3, 1977. EPAs legal authority to require greenhouse gas emissions reporting from
the petroleum and natural gas sector falls squarely within EPAs expansive statutory authority to
collect information under sections 114 and 208 of the CAA and is consistent with past EPA
practice.
In short, we have already lost a years worth of critical emissions data from the petroleum and
natural gas sector. EPA has both a legal duty to promulgate rules requiring the collection and
reporting of this data and broad statutory authority supporting promulgation of such rules. We
appreciate EPAs reaffirmation in this supplemental proposal of the urgent need to require
reporting from the petroleum and natural gas sector, and we urge EPA to move expeditiously in
finalizing this rule, as well as the reporting requirements for other sectors missing from the
October 2009 Rule, in order to satisfy its legal mandate to require emissions reporting from all
sectors of the economy of the United States.
Response: EPA initially proposed Subpart W in 2009 with the other sectors of the industry.
However, stakeholders raised concerns regarding the burden to report to the 2009 proposal as
970
well as the emissions coverage that would be missing due to the exclusion of onshore production
and distribution sectors from the 2009 proposal. To address these issues, EPA re-proposed
Subpart W in 2010 and has today finalized the rule.
EPA notes that the establishment of reporting requirements for these source categories is
appropriate and within EPAs legal authority. Please see the preamble (section II.Q) of The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), as well as Vol. 9
Response to Comments, Legal Issues on the 2009 final rule, for a discussion of EPAs legal
authority to collect the data required by this rule. See also 75 FR 39736, 39752 (July 12, 2010)
for a discussion on how the appropriations language grants EPA much discretion to determine
the appropriate source categories to include in the reporting rule. On a national basis, GHG
emissions from the oil and gas industry are second only to power plants. Hence, EPA has
determined that it is necessary to collect an accurate amount of GHG emissions information from
this sector.
971
VOLUME 10: COST AND ECONOMIC IMPACTS OF THE RULE
10.0
COST AND ECONOMIC IMPACTS OF THE RULE
No comments received.
10.1
METHODOLOGY USED TO ESTIMATE COST IMPACTS
Comment Number: EMAIL-0001-8 (comment also located in rulemaking memo Early
Comment Submissions in docket EPA-HQ-OAR-2009-0923)
Organization: American Exploration and Production Council
Commenter:

Economic Impact Analysis
We are in the process of evaluating the economic impact analysis provided by EPA. At this point
it appears that EPAs cost estimates for impact on the industry are low by one to several
magnitudes. EPA fails to realize the massive amount of data collection and recordkeeping
required to calculate emissions from the vast majority of some 700,000 well sites in the U.S. The
only feasible alternative is to adopt simple threshold exemptions and simplify the data gathering
requirements and calculations by use of widely accepted methodologies such as the API
Compendium. Industry and the EPA would be much better served by spending resources to
improve any emission factor methods that EPA feels is not representative of actual emissions
(e.g., well blowdowns).
We are particularly concerned with the requirement to count all fugitive components on every
well site and small production facility that we operate. This requirement is very significant in
terms of effort and cost. Thus, it is critical to us to receive relief on fugitive component
inventory.
Response: EPA has reviewed its estimates and disagrees that they are too low by an order of
magnitude or more. See EPAs response to comment EPA-HQ-OAR-2009-1151-107 for more
details. EPA also disagrees that emissions will have to be calculated from the vast majority of
700,000 wells in the U.S. See response to comment EPA-HQ-OAR-2009-0923-1151-89 for
further details.
EPA has considered the commenters recommendation for threshold exemptions and simplified
data gathering requirements. In particular, EPA agrees that the burden for counting components
for equipment leak emissions could be large to some reporters. Todays final rule has several
modifications that will considerably reduce reporting burden and simplify monitoring methods.
Please see Section II.E of the preamble (EPA-HQ-OAR-2009-0923) for complete details about
changes EPA has made in todays final rule to reduce the burden to quantify equipment leaks;
see Equipment-Level Population Emission Factors for Onshore Production (EPA-HQ-OAR
972
2009-0923) and EPAs response to EPA-HQ-OAR-2009-0923-1151-107 for more information

on quantifying emissions from major equipment counts. Please also see response to EPA-HQ
OAR-2009-0923-1151-5 for discussion about changes EPA made to todays final rule to reduce
burden for onshore production, such as requiring operators to count their major equipment
instead of each individual component.
Finally, todays final rule provides equipment thresholds that will significantly reduce burden to
monitor and report emissions. See Section II.F of the preamble and the response to comment
EPA-HQ-OAR-2009-0923-1151-5 for more details.
Organization: Southwest Gas Corporation
Commenter: Jim Wunderlin
Total Cost for Process Emissions
Given the above discussions, Southwest believes that the Process Emissions cost estimate
described in the preamble of the proposed rule is significantly underestimated. The published
cost estimate for the natural gas distribution sector is estimated to be $1,680,000 for start up and
$600,000 for annual costs thereafter. Southwest respectfully suggests that these costs are more
likely to be $1,680,000,000 and $600,000,000, respectively
Response: EPA has reviewed the cost estimates of the April 2010 proposed rule found in docket
(EPA-HQ-OAR-2009-0923-0002) and disagrees that the Agency significantly underestimated
the costs for the natural gas distribution segment. See EPA-HQ-OAR-2009-0923-1020-6 for
EPAs response to the commenters alternative cost estimate.

Organization: American Gas Association
Commenter: Pamela A. Lacey
EPA Has Vastly Underestimated Costs Changing or Eliminating the Annual Leak Survey
Proposal Would Help Reduce Compliance Costs Substantially
1. Reasons for EPAs Underestimate of Cost Impacts
EPA has estimated that the total industry-wide annual cost of complying with Subpart W for
natural gas distribution will be $1.6 million in the first year and $1 million per year thereafter. 98
EPA estimates that 143 LDCs will be subject to Subpart W using a facility threshold of 25,000
98
75 Fed. Reg. at 18625, Tables W-5 and W-6.

973
tpy CO2e. 99 Accordingly, EPA is estimating a per company cost of $11,188 in the first year and
about $7,000 per year per LDC thereafter. In contrast, as described later in these comments,
AGA members estimate the cost of complying with the leak survey requirement in the rule will
be orders of magnitude higher.
EPA has seriously underestimated the cost of complying with the 2010 Proposal, for at least four
reasons. First, the 2010 Proposal uses the undefined terms above ground meter regulators in
sections 98.232 and 98.233 100 and metering stations and regulating stations in section 98.238
(defining a natural gas distribution facility) that could be interpreted to include not only city
gates and large custody transfer or district metering and regulating (M&R) stations but also
industrial, commercial and even residential customer regulating and metering equipment. We do
not believe this was EPAs intent, but unless the term is clarified, regulatory uncertainty and the
risk of varying interpretations by field enforcement personnel would drive LDCs to include all
customer meters in their leak surveys and reporting. When multiplied by 65 million customer
meters across the country, the significant annual leak survey costs would result in billions of
dollars of unnecessary cost to gas utilities and their customers.
Second, the agency has underestimated the cost of conducting an annual leak survey of the eight
listed components using optical scanning equipment at distribution city gates, above ground
meter regulator stations and at underground storage facilities. Requiring the use of infrared
cameras significantly increases this cost.
Third, our members do not have inventories or records for all eight of the components listed in
section 98.232(f)(5) and (i)(1). They would have to visit each meter and regulator location
(depending on the definition of M&R) to develop the list required for applying component-level
emission factors to component counts. The burden of this proposal could be reduced significantly
by reducing the number of components and better defining them so it is clear what is to be
included. The burden could be eliminated by allowing the use of facility-level emission factors
for city gates and above ground M&R facilities, as EPA has proposed for below-ground M&R
facilities and vaults.
Fourth, EPA has underestimated the costs of making an initial threshold determination for small
distribution systems and other small facilities that likely fall below the threshold. These facilities
apparently would have to conduct a full leak survey using optical scanning in the first year in
order to determine whether a distribution, underground storage, transmission compression
facility or LNG facility does or does not exceed the 25,000 tpy regulatory threshold under
98.2(a)(2). EPA could avoid this burden by allowing small facilities to use a simpler threshold
determination method that does not require leak surveys or other field work.
Because M&R is Not Clear Defined and Could Apply to Customer Meters, Distribution M&R
Leak Survey Costs Could be Extremely Burdensome and Costly
99
EPA, Economic Impact Analysis for the Mandatory Reporting of Greenhouse Gas Emissions Under Subpart W
Supplemental Rule (GHG Reporting), Final Report (Dec. 8, 2009) (Economic Analysis) at page 4-5.
100
Proposed 40 C.F.R. 98.232.(i), 98.233(q)(7), 75 Fed. Reg. 18637, 18643.

974
Even without requiring infrared cameras, EPAs annual leak survey requirement would impose
far greater costs than it has estimated, largely because the proposal could be interpreted to apply
to millions of customer meters. The cost of employee time for visiting an individual meter
installation is not large, but when multiplied by millions of meters, the costs expand
exponentially. Based on a recent survey, discussed in greater detail below, AGA members
estimate that the incremental cost for conducting the annual leak detection called for in Subpart
W would be roughly $40 per inspection per city gate station or district regulator station or per
industrial or commercial customer meter if infrared cameras are not required roughly one hour
of technician time plus travel time.
Response: EPA disagrees that the Agency underestimated the rules compliance costs by orders
of magnitude. Please see Section III.B of the preamble to todays final rule for EPAs response
to three of the assumptions that the commenter identified above as the basis for its alternative
estimate (inclusion of customer meters; annual leak survey costs; and cost to make reporting
determinations, e.g. threshold determinations). As explained in Section III.B of the preamble,
EPA has identified the specific assumptions that accounted for the discrepancy between the
estimates; EPA also updated its estimate to include the cost for non-reporters to make a reporting
determination.
EPA agrees that the costs to visit an individual meter installation are not large, but disagrees with
the commenters total estimate because it included many sources that are not subject to todays
final rule. The requirements do not apply to customer meters (industrial, commercial, and
residential meters) or to farm taps. See preamble Section II.F for a complete discussion about
the meters. EPA intended to include in the rule. In addition, EPA has revised todays final rule
to allow alternative options for leak detection. See preamble Section II.F for details about the
alternative leak detection options.
Regarding the inventory of components, EPA agrees that counting all the components, regardless
of whether they are leaking, at each facility is burdensome. Therefore, todays final rule requires
reporters to inventory leaking components only at above ground custody transfer city gate
stations. Furthermore, todays final rule clarifies that the requirements do not apply to customer
meters (industrial, commercial, and residential meters) or to farm taps; see preamble Section II.F
for a complete discussion about the meters EPA intended to include in the rule.
Regarding the costs to inventory leaking components, EPA included travel to custody transfer
city gate stations and the associated labor in its estimates. In todays final rule, EPA assumed the
following for each of these stations: 30 minutes to complete a station survey; per diem costs of
$50 and accommodation costs of $100 per day; and 30 minutes driving between surveyed
stations at a cost of $0.50 per mile traveled.
Because todays final rule does not require a component count at non-custody transfer M&R
stations, underground vaults, and main and service lines, EPA did not assume travel costs to
these locations. Instead, reporters can estimate emissions for these sources using population
emission factors. Therefore, EPA estimated the costs for reporters to apply emission factors to
above ground meter and regulators (non-custody transfer city gate stations) as well as below
975
ground meter and regulators and vaults that are not customer meters. Furthermore, EPA did not
estimate any costs for customer meters because they are not subject to the rule.
EPA agrees that additional clarification and only the use of population emissions factors (i.e.
facility-level emission factors) would reduce compliance costs. However, EPA determined that
the use of leaker emissions factors (i.e. emission factors applied to components that are found to
be leaking during a leak survey) will provide an estimate of actual emissions, as opposed to the
use of population emissions factors, which estimates the "potential of emissions from each
station (see Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background TSD (EPA-HQ-OAR-2009-0923-0027), Section 4). EPA determined that
estimating actual emissions from equipment leaks in the processing, transmission,
underground storage, LNG storage, LNG import/export terminals, and distribution segments is
necessary to track changes in emissions from the various facilities and effectively inform policy.
Therefore, EPA did not modify the requirement to use leaker emission factors at above ground
meter and regulator city gate stations at which custody transfer occurs.

Commenter: James F. Wunderlin

Cost of Component Inventory
Unlike many utilities, Southwest has a robust electronic work management system which could
be used to track components that emit GHGs. Unfortunately, the types of components required to
be inventoried for compliance with the proposed regulations are not included in this electronic
data base and it would be very costly and time consuming to conduct the initial inventory.
Furthermore, the cost of conducting a survey of regulator stations during the required five-year
cycle is approximately $2.7 million. If an annual inventory is required for all regulator stations,
the five-year cycle cost would be approximately $9.7 million. These costs do not include
inventories of single-family residential meters and regulators.
Response: EPA considered this comment and similar statements from other commenters
regarding the costs to inventory equipment.
EPA is unable to evaluate the commenters estimates ($2.7 million and $9.7 million) because
insufficient documentation was provided to explain how these figures were calculated. For
example, the basis for the commenters estimate that a component inventory would incur a fiveyear cycle cost of $9.7 million is unclear, given that the commenter did not specify assumptions
about time and activities to inventory. Nonetheless, EPA has inferred that a large part of the
commenters cost estimate can be attributed to the misinterpretation that customer meters are
subject to leak detection requirements. This rule, however, does not require reporters to count
the leaking components from all stations and meters. See EPAs response to comment EPA-HQ
976
OAR-2009-0923-1016-11 for a discussion that clarifies the component count requirements under
todays final rule and for discussion about how EPA account for the labor and time involved in
the inventory.
While the commenters electronic work management system may not include all types of
components that are prescribed for monitoring under subpart W, EPA can not possibly develop a
rule that could match each and every companys work management system and EPA does not
prescribe that the commenter must use this system in relation to Subpart W. However, EPA
disagrees with the commenters assessment of the burden for the distribution segment because
todays final rule requires component leak detection only at above ground meter & regulator city
gate stations at which custody transfer occurs (this does not include customer meters). EPAs
response to comment EPA-HQ-OAR-2009-0923-1016-11 clarifies the requirements and
associated costs.
Commenter:
General Comments Cost of Implementation
EPA has grossly underestimated the cost of implementation for this reporting rule. According to
EPAs guidance, staff activities and labor were taken into consideration when drafting the
proposed rule. However, if the required sources to report remain aggregated by basin, the
number of affected facilities as well as travel to those affected facilities in remote areas does not
appear to have been taken into consideration; nor has the GHG emissions from vehicles traveling
to remote locations been taken into consideration. Furthermore, the EPAs cost estimation
assumes that staff exists on-site to assist with GHG reporting implementation: the opposite is
true. Many operators will be forced to hire additional temporary staff or utilize consultants to
review and compile data from various divisions in order to determine if Yates is indeed in
compliance with the proposed rule.
Response: EPAs cost estimate accounts for the labor and equipment costs expected to occur as
a result of this rule; EPA did not include costs incurred in the absence of this rule. EPA
disagrees with the commenters characterization of its cost estimates. In particular, EPA did not
assume staff exist on-site to assist with GHG reporting implementation. Furthermore, operators
will not need to hire additional staff or make special trips to comply with Subpart W, as
explained below.
EPA disagrees with the commenters implied assumption that facility workers do not visit well
sitesremote or otherwiseat least once per year. While EPA recognizes that well sites
generally operate unmanned throughout the year, operators still monitor the associated
equipment of interest under todays final rule, such as gas-liquid separators, wellhead
dehydrators, wellhead compressors, stock tanks, and heaters. For example, operators monitor the
wellhead compressor lubrication and operating hours, glycol dehydrator glycol regenerator
temperatures, glycol chemical color and level, glycol circulation pump flow setting, and
methanol inventory and injection pump rate settings. Also, company maintenance practices
977
require operators to periodically visit well sites that have this operating equipment to ensure, for
example, that fluid levels are proper and that automatic equipment functions properly; in the case
of well venting for liquids unloading, this is often a manual operation. Depending on the
wellhead equipment, this site visit may be once per year or, for wells that need frequent manual
liquids unloading, as frequently as once per week. These visits will take place even in the
absence of todays final rule.
However, EPA agrees that counting individual components at well sites is burdensome and has
revised todays final rule to allow reporters to instead count major equipment to quantify
equipment leaks. Operators will often know major equipment counts because they keep an
inventory for maintenance purposes. In addition, EPA agrees that many well sites operate
without any significant wellhead equipment. Given that todays final rule requires a count of
major equipment rather than individual components, operators will not need to visit wells
without major equipment at the wellhead
For the wellsites with major equipment, operators can count their major equipment during
routine visits to well sites even if they dont already have this inventory available. Accordingly,
EPAs burden estimate included the extra time spent at each site to collect relevant data for
reporting to this rule but did not account for the travel to site because it would occur in the
absence of this rule. For more information on quantifying emissions from major equipment
counts, please see rulemaking docket (EPA-HQ-OAR-2009-0923) under Equipment-Level
Population Emission Factors for Onshore Production.
Finally, the commenters did not provide sufficient information about its recommendation to
account for the GHG emissions from vehicles used to travel to well sites. Based on EPAs
interpretation of this statement, the recommendation irrelevant and beyond the scope of the
economic impact analysis. If the commenter was referring to emissions reported under Subpart
W, EPA notes that the rule does not require reporting of vehicle emissions. Rather, Subpart W
requires reporting of emissions data about key sources in the petroleum and natural gas industry.
Alternatively, the commenter may have intended to suggest that EPA evaluate the cost of
impacts from vehicle GHG emissions rather than the cost to monitor them. As discussed in
chapter 1 of the EIA, the purpose of the economic impact analysis is to illustrate the types of
costs and benefits that may accrue as a result of Subpart W, in particular the compliance costs for
reporters. EPA does not expect the monitoring activities under Subpart W to significantly affect
the level of GHG emissions from each reporters facility, since most of the well sites are already
being visited as a part of routine operations. Second, even if EPA were to estimate the GHG
emissions associated with Subpart W compliance activities, it would be incorrect to attribute all
vehicle emissions resulting from travel to well sites to Subpart W because such travel occurs in
the absence of this rule.

Organization: Pioneer
Commenter: Gretchen Kern
978

EPA's estimate to comply with Subpart W is greatly disproportionate to Pioneer's calculated
estimate and compliance will be unduly burdensome
Pioneer believes the EPA has severely understated the cost and burden of this Subpart.
According to EPA documents, EPA has estimated that it will cost an operator $ 18,000/facility
for the first compliance year, and $8,000/facility every year thereafter. As stated above, the cost
Pioneer has calculated for laboratory work to conduct sampling in the Permian Basin alone as
mandated by the Onshore Production segment, $17.2 million, greatly exceeds the EPA estimated
costs and will be an extremely enormous burden from a personnel and cost standpoint. This cost
does not include compliance with 98.233 (u) and (v), mentioned above, to determine the mole
percent in order to calculate the volumetric and mass emissions for CO2 and CH4 for many of the
source categories; these sections were very unclear and it was impossible to determine the
sampling protocol required. It also must be emphasized again that this cost estimate is only for
one of Pioneer's many asset areas, Permian. Further, it does NOT include the cost of manpower
to travel to each battery, count each of the 21 equipment sources potentially present at each
wellhead and battery (as mandated by the Onshore Production segment of this Subpart, take the
samples, or enter the data. Nor does it take into account the equipment required for sampling or a
GHG software program to manage the data as explained in point 9.
Response: EPA is unable to evaluate the commenters estimate [of $18,000/facility and
$8,000/facility] because insufficient documentation was provided to explain how these figures
were calculated. Regarding the EPA estimates cited by the commenter, they refer to the cost
($18,000) for the average operator per facility for the first compliance year and $8,000 per
facility for every year there after; larger operators and facilities will incur greater costs
depending on geographic dispersion and equipment owned. The average cost for a large onshore
production operator per facility is $35,000. Please see the response to EPA-HQ-OAR-2008
0508-1151-107 for a discussion on estimating Subpart Ws burden on the onshore production
segment.
Regarding the comments about sampling, please see EPAs response to EPA-HQ-OAR-2009
0923-1011-21 for more details on Subpart Ws sampling requirements and Section II.F of the
Preamble (EPA-HQ-OAR-2009-0923) for EPAs response to this comments and others
concerned with sampling burden.
Finally, EPA disagrees with commenter that EPA has not accurately accounted for the costs to
visit well and ancillary well sites and the associated monitoring (count components and take
samples). Please see EPAs response to EPA-HQ-OAR-2009-0923-1060-1 for how EPA
accounted for well site and ancillary well site visits and response EPA-HQ-OAR-2009-0923
1151-106 about the use and cost of GHG software programs.
.
979

EPA has greatly underestimated the cost and effort to conduct the direct measurement prescribed
for these sources. Based on the monitoring requirements of the proposed rule for the
reciprocating compressor rod packing venting source type alone (only one of 21 source types
listed under the onshore petroleum and natural gas production sector), an IPAMS member
company projects that the cost will exceed $108,000 per compressor unit alone and in excess of
$650,000 per facility (as EPA has defined the term for a hydrocarbon basin) based on the
following cost data and assumptions:
This is in comparison to EPAs Subpart W estimate of $18,000 for all source types combined at
an entire facility for the first year. Any one company will have hundreds or thousands of
compressors in operation, each one with multiple compressor rod packings, and many that may
be leased rather than owned by the reporting entity. Additional complexity and cost are presented
in the many cases when a single compressor is used for multiple services, for example inlet
compression, residue compression, and refrigeration compression. To further complicate the
required reporting, EPA has required that the GHG measurements be conducted three times for
each unit: once while operating, once while in standby mode, and once when depressurized. Not
only does this triple the already onerous resource requirements to collect the data, but requires
the depressurization of a compressor, thereby creating additional GHG emissions via compressor
blowdowns, just to take a third measurement that will likely be a zero reading unless a valve
has malfunctioned. Therefore, IPAMS requests that EPA allow operators to use the API
Compendium of Greenhouse Gas Emission Methodologies for the Oil and Gas Industry (API
Compendium) to calculate GHG emissions from rod packing venting.
Response: In response to this comment and others, EPA has decided not to require direct
measurement of reciprocating rod packing vents for onshore production under todays final rule.
Please refer to :Compressor Modes and Threshold (EPA-HQ-OAR-2009-0923) for the basis for
this change and details on quantifying emissions from reciprocating compressor rod packing. In
addition, todays final rule clarifies that EPA did not intend for compressors to be taken offline
in order for reporters to collect the data required under Subpart W. However, EPA disagrees
with the recommendation to use the API Compendium; please see comment EPA-HQ-OAR
2009-0923-1061-14. See EPA-HQ-OAR-2009-0923-0055-16 for complete discussion about the
monitoring requirements and compressor mode and the monitoring requirements.
980
10.2 VERIFICATION METHODOLOGY AND COSTS

This is important, because the type of leak detection equipment required by Subpart W could
affect the cost significantly. The following are general estimates of the cost of each of the
possible devices that could be considered to be optical gas scanning equipment -FLIR camera/ Optical Imaging Device: Cost around $80,000 to $100,000 each to purchase
around $2,500 per week to rent;
Mobile or Remote Methane Leak Detector (RMLD): Cost around $30,000 each;
Vehicle-Mounted Methane Leak Detector: Cost around $16,000-18,000 each; and
Remote hand-held optical methane detectors: Cost around $8,000 each.
In Exhibit B to these comments, we provide estimated costs of conducting the proposed annual
survey using optical scanning devices assuming this term refers to infrared cameras. AGA
estimates that it would cost about $2,900 per city gate to conduct the annual inspection using an
infrared camera, compared to $40.51 per city gate using other, often more methods that are more
effective for finding small leaks and are currently used by LDCs in leak surveys required under
existing federal and state pipeline safety regulations. The costs of using infrared cameras at other
M&R locations varies by type of metering location
- $2,900 per city gate
- $860 per district M&R station from transmission (custody transfer other than at city gate)
- $244 per district M&R station (stepping down pressure within distribution system)
- $377 per industrial meter
- $30 per commercial meter
- $30 per residential meter (assuming similar to commercial meter costs)
One can multiply the per location estimates by the total number of city gates, custody transfer
stations, industrial meters, commercial meters and residential meters in the U.S. (based on the
DOE EIA numbers of meters) to determine total cost for requiring infrared cameras at each type
of metering and regulating location. This will result in a conservative estimate of cost impacts, as
this member did not include travel time for the service personnel and did not use fully loaded
labor costs (i.e. did not include the cost of benefits).
AGA conducted an informal survey of its membership in June 2010 to estimate the number of
city gates, district M&R stations, and industrial, commercial and residential meters that are
potentially impacted by the proposed rule. AGA members reported the following:
981
- Each LDC serves an average of 143 city gate/custody transfer stations

- Each LDC serves an average of 1,434 district regulator stations on distribution lines and 118
district regulator stations on intrastate transmission lines
- Each LDC serves an average of 9,498 industrial meters on distribution lines and 256 industrial
meters on intrastate transmission lines
- Each LDC serves an average of 57,616 commercial meters
- For LDCs that track multi-family residential meters separate from commercial meters, each
LDC serve an average of 265,878 multi-family residential meters
- Each LDC serves an average of 789,758 residential single-family meters
From these estimates, one can calculate that each LDC operator will, on average, incur the
following annual costs if required to use infrared cameras for leak detection surveys:
- $413,250 for city gates
- $101,193 for district M&R stations from transmission (custody transfer other than at city gate)
- $349,877 for district M&R stations (stepping down pressure within distribution system)
- $3,677,581 for industrial meters
- $1,728,468 for commercial meters
- $7,976,346 for multi-family residential meters
- $23,692,732,181 for single-family residential meters
These costs are absurdly high, especially considering that our members report that the infrared
FLIR cameras often cannot locate the typically small leaks in distribution systems that can be
found using a simple soap bottle leak detection technology. This and other traditional methods
may be less sexy but they are usually far more accurate and less expensive that infrared FLIR
cameras for finding small leaks. It would be truly absurd to require an LDC to pay $100,000 for
each infrared FLIR camera, when they can get better results using a bottle of soapy water for a
few dollars. The optical scanning equipment should be an option but not a requirement.
This informal survey also collected estimated costs for leak inspection that does not include
infrared camera costs. AGA members estimated the following leak inspection costs per LDC per
year, on average:
- $33,938 for city gates
- $281,390 for district M&R stations
- $1,196,977 for industrial M&R station
- $4,603,376 for commercial meters
- $19,235,507 for multi-family residential meters
- $15,761,135 for single-family residential meters
AGA estimates that the requirement to use infrared cameras for leak detection will cost each
operator, on average, $37,939,450. In addition, AGA estimates the additional resources that
would be needed to fulfill the leak inspection requirements in the proposed Subpart W will cost
each operator, on average, $41,112,323 per year. The total potential fiscal impact for each LDC
operator is, therefore, $79,051,773 per year. These costs reflect the first year of implementation.
The following years would require the same level of resources of leak inspection, but would
require a different level of resources for the use of infrared cameras. These costs would depend
982
on the purchasing or renting of infrared cameras, the training of personnel using this technology
and the administrative and maintenance costs associated with the implementation of this
technology.
These fiscal estimates reflect responses provided by 30 LDC operators with widely varied
system compositions and operational considerations. A more accurate cost estimate would
require additional time that the comment period does not allow. However, from the responses
provided, AGA is comfortable asserting that the strain the 2010 Proposal would place on limited
resources is overly burdensome. Given the low emissions from distribution systems, clearly, the
cost per ton would be excessive even for a cap and trade system, let alone a reporting rule that
will not result in any GHG emission reductions.
There are more than 70 million residential, commercial and industrial customers in the U.S. If
leak surveys were required at each meter, even without using infrared cameras, the cost would be
$2,835,700,000 nearly $3 billion per year, and that is without requiring infrared cameras.
AGA accordingly requests that EPA allow the use of other effective leak detection equipment
and to revise sections 98.232, 98.233 and 98.234 to clarify that the term Natural Gas Distribution
Facility does not include customer meters and that LDCs are not required under Subpart W to
leak survey or report fugitive emissions from industrial, commercial and residential customer
meters.
EXHIBIT B to AGA Comments, June 11, 2010, Docket No. EPA-HQ-OAR-2009-0923
GHG Reporting Sub Part W Potential Impacts
983
Table 1
Impacted Facilities
Number ofFacilities
CostForAdditional
Annual Inspection
Cost For Optical

Scanning/Infrared
Cameras
Cost Inspectionduring Cost for Using

Cost For Additional
Existing Leak
Administration/ Record Normal Cycles
Detcection
Retention
Equipment
City Gate Stations

Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
4
91
179
38
0
0
Subtota l
$ 162
$ 3,686
$ 7,251
$ 1,539
$
$
$ 12,639
$ 64,500
$ 387,000
$ 129,000
$ 129,000
$
$
$ 709,500
$ 6,077
$ 3,929
$ 4,051
$2,026
$ 129,000
$ 64,500
$ 64,500
$64,500
$ 16,082
$ 322,500
$ 32,084
$ 67,004
$ 35,649
$ 9,641
$ 74,376
$37,917
$ 193,500
$ 774,000
$ 193,500
$ 193,500
$ 967,500
$967,500
$256,671
$ 3,289,500
N/A
N/A
N/A
N/A
N/A
N/A
District Regulator Stationsat Pipeline

Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
150
97
100
50
Subtota l
District Regulator Stations
Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
792
1654
880
238
1836
936
Subtota l
1FTE $60,000
1FTE $60,000
.5FTE $30,000
.5FTE $30,000
.5FTE $30,000
.5FTE $30,000
Industrials Meters / Servedby

transmision pipeline
Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
9
77
10
6
8
0
$ 729
$ 6,239
$ 810
$ 486
$ 648
N/A
N/A
N/A
N/A
N/A
N/A
$ 8,912
Subtota l
All otherIndustrial Meters
Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
342
749
77
205
598
110
Subtota l
$ 17,318
$ 37,927
$ 3,899
$ 10,381
$ 30,281
$5,570
$ 129,000
$ 193,500
$ 64,500
$ 129,000
$ 193,500
$64,500
$105,377
$ 774,000
Commercial M&R Stations

Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
7,000
7458
1730
614
7026
1530
$212,678
$226,593
$ 52,562
$ 18,655
$213,467
$46,485
1FTE $60,000
1FTE $60,000
.5FTE $30,000
$770,439
Subtota l
Residential Customer
Sta te A
Sta te B
Sta te C
Sta te D
Sta te E
Sta te F
755000
1350000
235000
111000
442673
104711
$2,340,242
984
$10,191,000
$390,000
Table 2
CostAnalysis
Citygate /DRS
Industrial onTrans
Industrial
Commercial
CostforOptical Scanners
Inspect/Document Inspect/Documen
1st Time
t 2ndTime
ScannersNeeded
Drive Time
0
60
25
15
$64,000
$60
$60
$60
$30
$30
$30
$30
$15
1per4,000Commercial
Response: EPA has reviewed this comment and disagrees that the cost estimates of April 2010
proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002) has been underestimated by
orders of magnitude. As discussed in response to EPA-HQ-OAR-2009-0923-1016-11, the
commenter based its estimated annual inspection cost to use an infrared optical imaging
instrument on an assumption that EPA required leak detection and calculation of emissions from
customer meters (i.e., industrial, commercial, and residential meters) and from M&R stations at
which custody transfer does not occur. The requirements, however, do not apply to customer
meters (industrial, commercial, and residential meters) or to farm taps. See preamble Section
II.F for a complete discussion about the meters EPA has included in todays final rule. The
commenters assumption therefore resulted in a much larger estimate of compliance costs.
Specifically, this assumption accounted for about 99% of the commenters total estimate, or
approximately $37.4 million of the $37.9 million estimated per facility. 101 EPAs response to
comment EPA-HQ-OAR-2009-0923-1016-11 discusses the costs associated with the meters
EPA intended to include in the rule. The remainder of this response discusses the detailed cost
estimates submitted by the commenter, as shown in this excerpt (EPA-HQ-OAR-2009-0923
1016-16).
While EPA finds the commenters estimates of the capital cost of individual pieces of leak
detection equipment to be reasonable, the commenter did not provide enough documentation or
information to explain the basis for the total annual inspection cost estimates of using infrared
optical imaging instruments. Likewise, the commenter did not provide enough documentation or
information to substantiate its estimate for the additional resources needed to fulfill the leak
inspection requirements ($41 million per LDC). However, EPA considered comments stating
that it would be more cost-effective to use alternative leak detection methods. EPA has revised
todays final rule to allow alternative options. See preamble Section II.F for details about the
alternative leak detection options and see EPA-HQ-OAR-2009-0923-1016-11 for discussion
about how this revision affects the costs.
101
Note: The commenter stated that the cost for single-family residential meters was $23 billion. However, based
on the cost per residential meter ($30) and the average number of single-family residential meters (789,758)
served by an LDC, as cited by the commenter, the $23 billion estimate seems erroneously too high by several
additional orders of magnitude. Using the numbers provided by the commenter results in $23 million, an order of
magnitude that is consistent with the commenters total estimate. Therefore, EPA has assumed this was a
typographical error and based its response on the corrected estimate.
985

Covered Entities (Facilities)
As part of our analysis of the wells per operator in each basin, API also analyzed the number of
operators/basin combinations (potentially covered entities) and the number of entities covered
under various per well emission assumptions. This analysis determined the following:
- The sum of all operators with operating wells in all basins is 21,744 rather than the 27,993
listed in Table W-2 of the rule preamble.
- As illustrated in the following table the number of covered entities/facilities based on grouping
within a basin varies widely based on the assumptions of per well annual emissions.
- Even at EPAs lowest annual emission estimate per well, APIs analysis indicates coverage of a
larger population of entities under the Basin entity threshold construct. Coupled with the lower
number of potentially covered entities from our analysis the percentage coverage is more than
double EPAs 4% estimate even at the lowest emission estimate per well.
API Requests
- EPA should reassess their covered entity analysis as part of a broader reassessment of their
economic impact analysis and incorporate realistic cost information in the economic impact
analysis, as the basis for the rule requirements and preamble
Response: EPA agrees with the commenter that the number of operators/basin combinations
should be lower, but disagrees with the commenters estimate of the covered entities subject to
reporting. EPA has reviewed its threshold analysis, based on the HPDI database (used in todays
final rule), and found duplicate operators within some basins. This duplication occurred mostly
because HPDI collects data from individual States that may record the same operators name
slightly differently, e.g. leading or trailing punctuation in the operator name. For example, Joe
Browns Oil Company and Joe Browns Oil Comp. operating in the same basin refer to the same
reporter but would have counted as two different reporters because the listed operator names
986
were slightly different. EPA identified 5,483 such duplicates and has since eliminated the
duplicate names from the database. The total number of operators/basin combinations used for
the analysis in todays final rule totals 22,510.
Regarding the number of entities likely to meet the reporting threshold, EPA disagrees with the
commenters alternative analysis. The commenter assumed that EPA developed an estimate of
annual emissions per well and applied that across operators and basins to determine the number
of reporters at the 25,000 Mt CO2e threshold. The commenter used EPAs estimated individual
well pad emissions from the Technical Support Document (TSD) found in docket (EPA-HQ
OAR-2009-0923-0027) for the April 2010 proposed rule. EPA estimated individual well pad
emissions to analyze coverage at different thresholds for a scenario in which the facility is
defined as a well pad. EPA does not agree that this well pad level estimate can be extrapolated
to operators/basin combinations. Extrapolating the well pad level estimates requires an inherent
assumption that the emissions per well are identical. EPA disagrees with this assumption
because the quantity and type of equipmentand therefore the magnitude of emissionsvary by
well. Moreover, this assumption is unnecessary because available production data are detailed
enough to facilitate analysis by operators. Please see the Section 6 of the Technical Support
Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923) for more
details on EPAs well pad threshold analysis, which uses a completely different methodology
than the basin level definition methodology. Also, see Section II.F of the preamble for details
about how EPA estimated emissions per well. Hence, EPAs estimate that the rule covered 4
percent of entities in the onshore production segment at the 25,000 Mt CO2e threshold is
appropriate and consistent with the qualified apportionment of emissions discussed in the TSD.

EPA must delay inclusion of the Onshore Petroleum and Natural Gas Production Sector in the
final rule pending a re-evaluation of the Economic Impact Analysis and rationalization of the
rule requirements, rule approaches, burden, and costs vs. the amount of emissions covered
Response: EPA has reviewed all comments on the Economic Impact Analysis and, where
appropriate, revised its estimates. See preamble Section III for a summary of the analysis and
the Economic Impact Analysis (EIA) (EPA-HQ-OAR-2009-0923) for complete results.
Todays final rule allows reporters to request the use of best available monitoring methods under
certain conditions, such as lack of service providers and qualified technicians; please see the
response to EPA-HQ-OAR-2009-0923-1011-27 for more information.
10.3
COST IMPACTS

987
The total estimated first year cost of compliance is approximately $16,000,000 with an average
of about $8.50/tonne CO2e. For subsequent years, the estimated total cost of compliance is
approximately $11,000,000 with an average of about $6.00/tonne CO2e. Although, as noted in
the table, emissions and/or cost estimates were not available for all the emission sources, it is
expected that over 90% of the costs are included. It should be noted that companies that have
different production field characteristics (e.g. well completions and workovers, compression and
dehydration requirements, gas-driven pneumatic device population) would have a different mix
of primary emission sources and different cost factors.
988
989
A review of the cost data in Table 1 shows:

EPA has drastically underestimated the cost of rule compliance.
- The total estimated first year cost of compliance for Noble Energy alone is approximately
$16,000,000 with an average of about $8.50/tonne CO2e. Table W-5 of the proposed
rule preamble estimates the first year cost of compliance for the entire onshore
production sector to be about $22,700,000 with an of $0.18/tonne CO2e. The EPA cost
estimate is about a factor of 50 lower than the Noble Energy estimate.
- The total estimated cost of compliance for subsequent years for Noble Energy alone is
approximately $11,000,000 with an average of about $6.00/tonne CO2e. Table W-5 of
the proposed rule preamble estimates the subsequent years cost of compliance for the
entire onshore production sector to be about $8,600,000 with an of $0.06/tonne CO2e.
The EPA cost estimate is about two orders of magnitude lower than the Noble Energy
estimate.
Very high compliance costs for numerous emission sources indicate alternative, simpler
emission estimation methods are needed or that these sources should be removed from the
reporting requirements.
- Annual costs for Hydrocarbon Liquids Dissolved CO2 are about $40,000/tonne CO2e.
These costs are a result of this being a very small emission source (as shown in Table 2
below) and the requirement for quarterly sampling of liquid hydrocarbon storage tanks.
As discussed below, this is an insignificant emission source to the total inventory.
- Quarterly sampling requirements contribute to the high costs for Produced Water
Dissolved CO2 and Acid Gas Removal Vent Stacks. As discussed above, AGRs are an
insignificant emission source to the total inventory.
- Extensive process sampling requirements contribute to the high costs for Production
Storage Tanks and Glycol Dehydrators.
- Surveying thousands of well sites and annual tracking of new, decommissioned, and
divested operations contribute to the high costs for Component and Gathering Pipeline
Fugitives. As discussed below, available data indicate Gathering Pipeline Fugitives is an
- Direct measurement requirements contribute to the high costs for Reciprocating
Compressors Rod Packing Vents and Well Venting for Liquids Unloading. As discussed
990
below, available data indicate Reciprocating Compressors Rod Packing Vents is an

Response: Regarding comments that EPA underestimated the costs for onshore production, see
preamble Section III.B.2 of todays final rule for EPAs response and discussion about the basis
for the cost estimates. See also comment EPA-HQ-OAR-2009-0923-1167-11 for EPAs
response to the specific cost estimates presented above. Regarding the commenters cost issues
with monitoring methods, see EPA-HQ-OAR-2009-0923-1151-5, which discusses changes EPA
made to the todays final rule to reduce burden for onshore production, such as requiring
operators to count their major equipment instead of each individual component.
3. Estimated Cost Impact on Oil and Gas Production
As described earlier, Devon has evaluated the economic impact of the proposed Subpart W. We
performed a source by source cost analysis using recent equipment counts. Devons cost estimate
for the first year of compliance is nearly $28 million, without contingency.
APIs projected industry-wide first year cost estimate exceeds $1.8 billion and is larger than
EPAs projection of $27.7 million by a factor of at least 66 times. Although API used a different
approach to evaluate costs, there is congruence between Devons and APIs cost projections
because we operate approximately 2% of wells in the US and our cost estimate comprises
approximately 2% of the cost that API estimates.
Devon is convinced that EPA has significantly underestimated the cost impact and burden of
Subpart W. Therefore, EPA should:
* Re-evaluate the Economic Impact Analysis and restructure the rule to simplify the
requirements and reduce burden.
* Delay inclusion of the Onshore Petroleum and Natural Gas Production sector. This could be
accomplished by developing a phased implementation schedule for Subpart W so that industry
sectors with less resource-intensive requirements start reporting sooner and those with more
burdensome requirements start reporting later.
* Allow use of best available monitoring methods (BAMM) for source types requiring metered
flow rates or monitoring parameters for calendar year 2011.
Response: EPA disagrees with the cost estimate provided by the commenter; please see
comment EPA-HQ-OAR-2009-0923-1174-2 for further details. Also, see comments EPA-HQ
OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923-1151-106 for detailed explanation
of EPAs analysis of APIs comments and conclusions that the commenter has cited. Regarding
991
the comment that EPA revaluate the Economic Impact Analysis (EIA) and restructure the rule,
see preamble Section III for a summary of the economic impact analysis conducted for todays
final rule. Regarding the commenters cost issues with the monitoring methods, see EPA-HQ
OAR-2009-0923-1151-5, which discusses changes EPA made to the final rule to reduce burden
for onshore production, such as requiring operators to count their major equipment instead of
each individual component. Finally, regarding the commenters recommendation to delay
implementation of the rule, see Section II.F of the preamble a complete response and discussion
about conditions under which reporters may use BAMM.
Preamble at 18618 and 18625.(KEY ISSUE) El Paso believes that EPA has significantly
underestimated the level of effort and cost implications for Subpart W, as it applies to onshore
oil and gas production operations. As a result, reporting should be based on a field basis. A
detailed cost analysis has been included in Appendix II.
Preamble at 18618 and 18625, Economic Impacts of the Proposed Rule Support Reporting on a
Field Basis as Opposed to Cost Prohibitive Basin Approach
El Paso supports commenters, such as the American Exploration and Production Council, who
urge EPA to use the traditional definition of facility for oil and gas activities, rather than either
the field-level or basin-level approaches proposed. As between field-level and basin-level
reporting, El Paso recommends that EPA employ the applicability threshold for exploration and
production based on applicable sources located within a reporting field or area that in the
aggregate emits over 25,000 metric tons per year. Through this rule development, in determining
the appropriate threshold and facility definitions, EPA must balance the coverage in terms of
emissions with the regulatory and cost burden to both industry and the regulatory agency.
Table W-2, on page 18618 of the preamble, indicates that at a proposed 25,000 metric tons
threshold, about 81% of the emissions are covered by employing the applicable facility defined
at basin level. However, as outlined in Table W-3 on page 18619, EPA concludes that about
55% of the emissions are covered by employing a facility definition at a reporting field level.
El Paso has been unable to reproduce EPAs emissions or cost basis due to lack of information
provided in various support documents associated with the rule. The preamble is the first public
document where EPA has revised its estimates of the total emissions from the natural gas sector.
EPA has not given reasonable consideration or explanation of the costs of including E&P in the
monitoring and reporting rule requirements.
El Paso believes that EPA has significantly underestimated the level of effort and cost
implications for Subpart W, especially as it applies to onshore oil and gas production operations.
However, it is difficult to compare internal estimates of the cost impacts to EPAs assumptions,
given that the Economic Impact Assessment (EIA), Technical Support Document (TSD), or
992
information available on the Docket are not consistent with the basin-level facility definition or
do not provide sufficient detail to evaluate EPAs cost assumptions. El Paso developed a detailed
cost estimate for our E&P operations and the number of sources associated with our upstream
operations for two reporting unit scenarios: (1) basin > 25,000 metric tons; and (2) field >25,000
metric tons. The detailed cost analysis has been included in Appendix II. Figure 2 presents
estimated costs for El Pasos onshore oil and gas production operations relative to the percent of
total emissions and number of regulated fields for each of the three reporting area scenarios.
Essentially, 98% of the emissions are covered at the basin level but at over twice the cost of
covering El Paso facilities at a reporting field basis.
Figure 2: El Pasos Estimated Compliance Costs for Three Reporting Area Scenarios [see
original PDF for figure]
A. Basin Reporting Unit >25,000 metric tons CO2e:
El Paso constructed a detailed cost analysis on a source-by-source basis for all of its operations
that exceed 25,000 metric tons CO2e at the basin level. For the upstream operations only, the
proposed rule will require that more than 1 million individual emission sources be measured or
documented in order to quantify and report emissions. El Paso has examined the level of effort
expected to gather data specified for onshore oil and gas production operations. Our estimates
show that first year compliance for Subpart W will require over 25 full time equivalents102
(FTE). Subsequent years will require 17.1 FTEs. Based on our estimates for the basin-level
reporting unit, the emission sources with the largest compliance burden are 10K thresholds.
- Dissolved CO2: 8.57 FTEs to comply (for both Year 1 and subsequent years);
- General activities to develop data management systems and processes (including training) to
gather, report, and monitor the required information: 4.58 FTEs for Year 1, 2.65 FTEs for
subsequent years.
- Tanks: 3.75 FTEs for Year 1, 3.21 FTEs for subsequent years.
- Gas well workovers and completions (conventional and unconventional combined): 1.08 FTEs
for Year 1 and 0.59 FTEs for subsequent years.
- Well venting for liquids unloading: 1.05 FTEs for Year 1 and 0.62 for subsequent years.
- Reciprocating Compressor Rod Packing Vents: 0.71 FTEs for Year 1, and 0.67 FTEs for
subsequent years.
For a reporting unit defined as a basin > 25,000 metric tons CO2e, El Paso has 9 such basin-level
reporting units comprised of 60 E&P fields, with an average first-year cost of $732,713 per
basin.
B. Field Reporting Unit >25,000 metric tons CO2e:
For comparison, El Paso also examined the costs and emissions coverage based on defining the
E&P reporting unit at a production field level emitting > 25,000 metric tons CO2e (these details
102
FTE based on 2000 hours/year.

993
are also provided in the cost analysis spreadsheet). This approach captures 65% of emissions
from El Pasos E&P operations, but reduces the reporting burden to a total of 15 production
fields at an average cost of $206,000 per field. Emissions associated with fields not covered by
the provisions of the proposed rule, could be estimated by applying simple emission factors to
production rates or estimates of major equipment.
Based on the field level approach and 25,000 metric tons threshold, the estimated first year
compliance will require 9.9 FTEs and 7.9 FTEs for subsequent year compliance efforts.
Table 2: Comparison of El Paso and EPA Cost Estimates
C. Reporting Level Conclusions and Recommendations:

Based on the detailed cost analysis performed using inventory information rooted in three years
of third-party verified voluntary emissions inventories, El Paso supports a production field level
approach and 25,000 metric tons CO2e per year threshold. The quality of emissions information
gathered would be better for field-level data rather than basin-level data. Characteristics of the
produced oil and gas, method of separation and processing are more closely aligned with the
field from which the oil and gas is produced than from a broader basin perspective. El Paso has
been compiling GHG emission inventories at the field level since 2006, using field designations
that are similar to that of the Energy Information Administration. We believe this approach
balances both coverage and regulatory and compliance burden.
APPENDIX II
Analysis of the EPA Proposed Greenhouse Gas Mandatory Reporting Rule Cost Impacts for
Natural Gas Systems
El Paso reviewed the EPA economic impact analysis for five types of natural gas industry
facilities affected by Subpart W of the proposed GHG mandatory reporting rule: (1) onshore
production facilities; (2) natural gas transmission facilities; (3) natural gas processing plants; (4)
underground storage facilities; and (5) LNG import facilities. The information made available to
the public through the rule docket is inadequate for conducting an independent review of EPAs
calculations and many of its assumptions that form the basis for the Subpart W cost estimates
given in the RIA for onshore production facilities. Nonetheless, El Pasos reviews indicates the
994
first year average cost impact 103 of $18,000 per pipeline transportation of natural gas system
facility and $24,000 per E&P facility is underestimated as shown in the following tables.
103
EPA Economic Impact Analysis for the Mandatory Reporting of Greenhouse Gas Emissions Under Subpart W
Supplemental Rule (GHG Reporting), Final Report, December 8, 2009.
995
996
997
Response:
Overview of Compliance Cost Estimates and Supporting Documentation
EPA disagrees with the comment and has determined that EPAs cost analysis accurately reflects
the compliance costs. Regarding the documentation, EPA has made all of the information
required to understand and replicate the Agencys cost estimates available to the public by
placing it in the docket (EPA-HQ-OAR-2009-0923). The Economic Impact Analysis (EIA)
998
serves as the core document; Section 4 of this document explains how EPA estimated the costs,
identifies and discusses underlying assumptions (e.g., number of labor hours assumed for each
activity and associated labor rates), and presents the detailed results and cost breakdowns. Also,
the EIA explains how EPA incorporated basin-level reporters in the cost model (Section 4.3),
breaks out costs specific to the basin-level reporting (see Tables 4-3 and 4-4), and discusses the
alternative field-level analysis (Section 5.1.4). The EIA specifies the basin-level approach used
for onshore production and Section 5.1.4 of the EIA shows how estimates at the field-level
compare with those at the basin-level. Additional, detailed supporting documents, such as the
spreadsheets underlying the threshold analysis and cost estimates for basin-level reporters in
onshore production, can be found in the docket; e.g., see Onshore Production Mandatory
Reporting Rule Analysis (EPA-HQ-OAR-2009-0923-0015). In sum, EPA does not agree with
the commenter that the information in the TSD, EIA, and other supporting materials in the
docket are not consistent with the basin-level facility definition. Rather, the EIA, TSD, and other
supporting materials document EPAs analysis at the basin-level for cost estimates.
Consideration of Commenters Onshore Production Cost Estimates: Field-level and Basinlevel
Comparing the compliance costs of a large facility, such as the one represented by the
commenter, to the average compliance cost from EPAs analysis is inaccurate and misleading.
While average onshore production compliance costs shed light on the costs from a national
perspective, they do not accurately reflect the likely costs for larger facilities. For todays final
rule, EPA estimated costs for each reporter using actual activity data derived from the HPDI
database. The activity data shows an exponential growth in the number of wells and production
for higher emitting producers. This means that a small number of reporters are large in size and
the rest of the reporters are considerably smaller in size. Consequently, the average cost for all
reporters is highly skewed towards the smaller facilities. In fact, the highest individual reporting
cost is estimated at $136,791 while the lowest individual reporting cost is estimated at $6,981.
This range shows the large potential difference in reporting costs among petroleum and natural
gas operators and why a simple average is not an accurate estimate for larger reporters like the
commenter.
Nevertheless the commenters estimated cost still exceeds EPAs maximum compliance cost
estimate. EPA has reviewed both the basin and field level cost estimates provided by the
commenter. However, the commenter did not provide sufficient details to substantiate the
estimate, such as assumptions relating to labor hours required for monitoring, to allow EPA to
respond in detail. The commenter provides FTE hours for monitoring each of the emissions
sources, but does not provide details about why that many FTEs would be required. Although
the commenter did not provide all the relevant details regarding their cost estimate, EPA has
reviewed the higher cost elements in the commenters total estimate and determined that the
commenter overestimated the costs for Hydrocarbon Liquids Dissolved CO2, Produce Water
Dissolved CO2, and Coal Bed Methane Produced H2O emissions, and Storage Tanks. EPA has
surmised that these gross overestimates may have resulted from unclear language in the proposed
rule and subsequent misinterpretation with regard to sampling requirements. Please see Section
II.F of the preamble for further details and EPA-HQ-OAR-2009-0923-1151-92 for a response to
a comment about the sampling requirements.
EPAs Analysis of Basin-level and Field-level Approaches: Emissions Coverage and Costs
Regarding the comments on the basin-level versus field-level reporting, EPAs analysis of
national emissions covered under both options shows that the field-level option would result in a
999
significantly lower coverage in emissions reported57 percent at field-level versus 85 percent at

the basin-level for a 25,000 metric tons CO2e thresholdand a higher cost per ton. Therefore,
EPA has concluded that basin-level reporting is more cost-effective because the cost per metric
ton to report under the basin-level definition is lower than under the field-level definition at the
reporting threshold. EPA notes that the total costs to report under the basin-level definition may
be higher but that is because a higher level of emissions are reported. See Section II.D of The
April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002) for more
information about EPAs decision to require basin-level reporting for the onshore production
segment.
Compliance Cost Estimates for Other Segments
Finally, for EPAs response to the commenters estimates for natural gas transmission facilities,
natural gas processing facilities, underground storage facilities, and LNG import facilities see the

Organization: Independent Oil and Gas Association of West Virginia
Commenter: Charlie Burd
To the extent that USEPA has concerns about the GHG emissions from those onshore production
wells that fall below the 25,000 tpy CO2e threshold, IOGA-WV notes that various estimation
tools are readily available to the agency to calculate these emissions without imposing
unnecessarily costly requirements on individual operators. In West Virginia, for example, the
West Virginia Department of Environmental Protection maintains a record of the number of
active oil and natural gas wells in the state, and operators of these facilities must monitor and
report their oil and gas production levels on an annual basis. In addition, as noted in EPAs
comments, USEPA operates the Natural Gas STAR Program, which the agency has already used
to identify emissions sources and estimate total emissions from the oil and gas sector. From this
information, and with the assistance of the various emissions estimation tools and methodologies
at USEPA's disposal, the agency should be able to estimate the marginal contribution from these
sources to the nation's total GHG emissions without singling out the oil and gas industry to bear
disproportionate costs of this proposal.
Response: EPA does not agree with the comment and expects todays final rule to provide the
public accurate emissions and inform future policymaking. The alternatives suggested by the
commenter, such as voluntary reporting or reliance on existing production data, would not
provide complete and comprehensive estimates of facility-level emissions. In particular, the
Natural Gas STAR Program does not provide sufficient emissions data to accurately characterize
emissions from the entire United States for several reasons. First, Natural Gas STAR Partners
represent just over half of the industry operators in terms of throughput. Second, the Partners
have usually targeted the most economic reduction opportunities, meaning the emissions
reductions reported are skewed towards the higher emitters. Finally, the Partners have not always
reported emissions before and after reductions.
Regarding selection of reporting threshold, see Section II.D of the preamble for more on the
threshold justification and monitoring requirements. In addition, Section 5 of the Economic
1000
Impact Analysis presents one of the analyses EPA conducted in considering the tradeoff between
monitoring costs and level of emissions coverage at various thresholds. For example, the
threshold cost-effectiveness analysis shows how the marginal costs and total emissions covered
change relative to the 25,000 ton threshold. Finally, EPA determined that screening tools would
facilitate reporting determinations and plans to make such tools available. Accordingly, EPA has
updated the Economic Impact Analysis to better account for reporting determinations and
expected use of screening tools; see EPAs complete response in Section III.B.2 of the Preamble
to todays final rule.
Regardless of how USEPA defines a "facility" for purposes of Subpart W, IOGA-WV has
concerns about the heavy burden that the initial applicability determination will have on
potentially affected oil and gas operators. Unlike Subpart C of the agency's mandatory GHG
reporting rule, 104 which applies to emissions from stationary fuel combustion sources, sources to
which proposed Subpart W may apply cannot conduct a relatively inexpensive, cursory review of
the number of units potentially covered by the rule and calculate whether or not monitoring and
reporting is required. Rather, USEPA's proposal forces oil and gas operators to conduct a year's
worth of monitoring on several systems in order to determine whether the rule is even applicable
in the first instance, using the same measurement and/or estimation methodologies that would
apply to future monitoring in the event that Subpart W ultimately is found to apply. Unlike other
industrial sectors, because oil and gas systems generally have not been subject to such extensive
monitoring requirements for their GHG emissions, many potentially covered entities do not yet
have GHG monitoring capability already in place. As such, determining whether Subpart W
applies---and, in the event that the rule does apply, implementing its various monitoring
requirements-will constitute a significant departure from current practice and will require the
investment of considerable time and resources. Indeed, in addition to the expenses associated
with the purchase and installation of this equipment, in many eases even assessing the
applicability of the proposed rule would require addition of staff to install, maintain and monitor
all of the systems covered by Subpart W. As discussed above, IOGA-WV has serious concerns
that the sheer expense associated with conducting this monitoring-even if only for purposes of
determining Subpart W's applicability-will result in the premature capping, plugging and
abandonment of these small or marginal wells, which could have devastating effects on well
owners, as well as employees, their families, and local communities that depend on them. The
omission of a less burdensome screening measure undermines the administrative and cost
benefits that USEPA has sought to achieve by selecting its 25,000 tpy CO2e reporting threshold
in the first instance.
104
See 40 CFR. 98.2(a)(3)(ii) (providing that those facilities containing stationary fuel combustion units, but no
source in any other source category covered by the mandatory reporting rule, are not required to report their GHG
emissions if their aggregate maximum rated heat input capacity from all stationary fuel combustion units is less
than 30 mmBtu/hr).
1001
For these reasons, IOGA-WV recommends that USEIA consider a capacity-based threshold or
other less costly emissions estimation methods or tools that would allow oil and gas operators to
determine whether the reporting requirements of proposed Subpart W apply. Such an approach
would simplify the applicability determination to minimize the burden of proving that a facility
is below the 25,000 tpy CO2e threshold. Any initial applicability determination methodologies
can be separate from those methodologies required for actual emissions reporting by covered
entities.
Response: EPA agrees that screening tools would facilitate reporting determinations and plans
to make such tools available. Accordingly, EPA has updated the Economic Impact Analysis to
better account for reporting determinations and expected use of screening tools; see EPAs
complete response in Section III.B.2 of the preamble to todays final rule. EPA does not agree
that reporters will require significant investment to comply with Subpart W, however. Please see
EPAs response to EPA-HQ-OAR-2009-0923-1151-107 for further details; EPAs response to
EPA-HQ-OAR-2009-0923-1151-5 discusses changes EPA made to todays final rule to reduce
burden for onshore production, such as requiring operators to count their major equipment
instead of each individual component; EPA has allowed the use of best available monitoring
methods for certain sources to further reduce burden, please see the response to EPA-HQ-OAR
2009-0923-1011-27 for more information; EPA-HQ-OAR-2009-0923-1005-7 discusses marginal
wells along with EPA-HQ-OAR-2009-0923-0053-1. Also, please see EPAs response to EPA
HQ-OAR-2009-0923-1303-2 for a discussion about Subpart Ws impact on the greater economy.
EPA has grossly underestimated the cost of implementation for this reporting rule. According to
EPAs guidance, staff activities and labor were taken into consideration when drafting the
proposed rule. However, if the required sources to report remain aggregated by basin, the
number of affected facilities as well as travel to those affected facilities in remote areas does not
appear to have been taken into consideration; nor has the GHG emissions from vehicles traveling
to remote locations been taken into consideration. Furthermore, the EPAs cost estimation
assumes that staff exists on-site to assist with GHG reporting implementation: the opposite is
true. Many operators will be forced to hire additional temporary staff or utilize consultants to
review and compile data from various divisions in order to determine if Yates is indeed in
compliance with the proposed rule.
In the Preamble (page 18610 of the Federal Register), the EPA states This proposed
supplemental rule incorporates a number of changes, including, but not limited to, different
methodologies that provide improved emissions coverage at a lower cost burden to facilities than
would have been covered under the initial proposed rule, the inclusion of onshore production and
distribution facilities While the original proposed rule called for reporting of 24 emission source
types, and this rule requires only 21, the EPA has lumped different types of pneumatic pumps
into more general categories, removed wet seal degassing and lumped fugitives into one general
category thereby resulting in a reduction in the number of reportable sources. Also, we concede
1002
that the original proposed rule required direct measurement of fugitives and other sources using
HIVOL sampling, bagging, etc However, the newly proposed rule includes reporting
requirements for many smaller sources that have limited or no emission data that were not
previously addressed as well as wellhead activities by basin and pipeline/gathering fugitives.
EPA is still requiring direct measurement of many of these sources, which is costly and time
consuming, and remains significant for operators to comply.
Response: EPA does not agree with the comment. While this commenter did not provide an
alternative cost estimate, EPA has analyzed other commenters cost estimates for onshore
production and determined that they are too high. A detailed analysis is available in EPAs
responses to comments EPA-HQ-OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923
1151-107. EPA-HQ-OAR-2009-0923-1151-107 identifies source categories discussed in each
comment.
Regarding the comment about source coverage, EPAs analysis shows that sources subject to
reporting under Subpart W release more than 80 percent of the emissions from the petroleum and
natural gas industry but comprise only 20 percent of total sources in the industry. Please see the
Technical Support Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009
0923) for further discussion on source selection. Nevertheless, EPA has responded to concerns
about the burden to inventory geographically dispersed equipment leak emission sources and
modified todays final rule requirements; see Sections II.C and II.F of the preamble (EPA-HQ
OAR-2009-0923) for a discussion of changes EPA made in monitoring requirements for onshore
production. Regarding commenters assumptions about travel costs and about hiring additional
staff or using consultants to determine if Yates is indeed in compliance with the proposed rule,
see EPAs response to EPA-HQ-OAR-2009-0923-1060-1. EPA has reduced the direct
measurement requirements for onshore production. Please see Section II.E of the preamble to
todays final rule for changes made to todays final rule. Also, please see EMAIL-0002-9
(comment also located in rulemaking memo Early Comment Submissions in docket EPA-HQ
OAR-2009-0923) for why EPA has retained direct measurement for some sources.
Commenter: Kim Dang
Comments on the Economic Impact Analysis. Kinder Morgan believes that EPAs Economic
Impact Analysis (EIA) of the proposed Subpart W significantly underestimates compliance costs
for the onshore petroleum and natural gas production source category and unrealistically assumes
that facilities with emissions less than 25,000 metric tons CO2-e will not have any compliance
costs. The effort for entities in the oil and natural gas sector to determine which facilities meet
the reporting threshold of 25,000 metric tons CO2-e per year is completely over looked. 105 Since
the proposed Subpart W provides no simplified or streamlined method for determining the
applicability of the rule to a given facility, Kinder Morgan would need to perform a full Subpart
W emissions assessment on almost all of its facilities in order to make certain that it is in full
105
40 C.F.R. SECTION 98.2(b) requires reporting facilities, including facilities that would be subject to the
proposed Subpart W, to apply the methodologies set forth in each Subpart to determine whether the reporting
threshold is met.
1003
compliance with the Mandatory Reporting Rule. For most facilities, such a screening assessment
may be required almost every year. Yet the EIA for the proposed Subpart W computes the
overall cost of the rule based on the assumption that 91% of facilities in the petroleum and
natural gas sector will be permanently excluded from the scope of the rule. 106 This is an
unrealistic assumption that results in the gross understatement of the cost of the rule. EPA cannot
assume the costs of making applicability determinations are zero. Even with a simplified
screening mechanism, there will be significant effort associated with measurements at those
facilities which are above any screening level and below the reporting threshold.
The industry structure and costs summarized in Tables W-9 and W-10 in the preamble of the
proposed Mandatory Reporting Rule use NAICS codes incorrectly. For example, onshore natural
gas processing plants fall under NAICS code 211112, and should not be lumped together with
onshore natural gas gathering compression under NAICS code 486210. LNG storage and import
and export equipment fall under NAICS code 486210, not crude petroleum and natural gas
extraction NAICS code 211. This mischaracterization of the industry structure through incorrect
application of NAICS codes may be diluting the onshore petroleum and natural gas production
cost impacts resulting in underestimating costs on a per entity basis.
For those facilities that are clearly above the reporting threshold, Kinder Morgan believes that
EPA has significantly underestimated the cost of the proposed Subpart W. The American
Petroleum Institute (API) performed a careful inventory of the costs associated with the proposed
Subpart W for the onshore petroleum and natural gas production sector alone, and determined
that this sector would incur a first-year cost of approximately $1.68 billion, about sixty times
EPAs estimate of $28 million. Building an inventory of well site components alone would cost
the industry $133 million, according to API. Of particular concern to Kinder Morgan is APIs
assessment that the proportion of onshore petroleum and natural gas production facilities covered
by the proposed rule is much higher than the 4% figure cited by EPA in the preamble to the
proposed rule.
Kinder Morgans own work with the proposed Subpart W has confirmed that the EIA cost
estimate is unrealistically low. Once the proposed rule was published, Kinder Morgan began a
series of field tests to explore the feasibility of implementing the proposed requirements. Based
on those field tests, Kinder Morgan has concluded that the cost per facility to conduct monitoring
in compliance with the proposed Subpart W for natural gas compression sector is over $14,000
per facility which includes the cost paid to contractors for labor and rental of the manlift and
other equipment. This burden would be incurred by Kinder Morgan alone for more than 200
facilities at a cost of over $3 million to determine applicability under the reporting threshold.
Importantly, this cost estimate reflects only ongoing year-to-year expenses associated with
Subpart W, and does not include the first-year costs associated with installing meters, training
personnel, and establishing data collection and quality control systems. In addition, EPAs cost
estimates do not appear to recognize that the installation of meters at many individual
components will require costly shutdowns of equipment. 107 EPAs cost estimate also fails to
106
EIA at p.4-10 (Table 4-6).
107
In part for this reason, Kinder Morgan recommends that EPA allow best available data to be supplied for the
first year of reporting (explained further below).
1004
evaluate the economic impacts that could result from service interruptions that could arise from
the requirement to monitor at three operational modes. Considering that EPA chose not to
finalize the original Subpart W in part because of the high industry burden it would have
imposed, 108 EPA must include these costs when considering if and how to finalize Subpart W.
Kinder Morgan notes that EPA appears to have erroneously applied NAICS codes to affected
facilities and sectors in Table W-10 of the preamble when calculating cost-to-sales impacts of the
proposed Subpart W. For example, natural gas processing facilities belong to the three-digit
NAICS code 211, whereas Table W-10 has classified them under NAICS code 486210 (pipeline
transportation of natural gas). It is also not clear whether LNG import, export, and storage should
be included in NAICS code 211 (crude petroleum and natural gas extraction), since these
activities are more closely related to natural gas transportation. Because the NAICS classification
of facilities in Table W-10 affects EPAs calculated cost-to-sales ratios for different segments of
the petroleum and natural gas industry, it is important for EPA to ensure that these facilities are
assigned the proper NAICS code in its economic analysis.
Even putting aside the costs not included in EPAs economic analysis, Kinder Morgan believes
the costs of Subpart W are still disproportionately high relative to other Subparts and industries.
Kinder Morgan notes that the first-year cost of the proposed rule ($0.21 per ton CO2-e across all
industry segments) 109 would be about seven times higher than the estimated first-year cost of
approximately $0.03 per ton CO2-e for the final Mandatory Reporting Rule as a whole. 110 In
subsequent years, EPA projects that the cost of reporting under the proposed Subpart W would
decline to $0.08 per ton of CO2-e 111 , which is still over twice as high as the first-year cost of the
Mandatory Reporting Rule. Given that even EPAs overly optimistic EIA found that the
proposed Subpart W would be disproportionately costly, Kinder Morgan urges EPA to consider
the above recommendations to minimize the costs of the proposed rule, including, for example,
establishing a screening tool to determine the applicability of the rule to individual facilities.
complete response in Section III.B.2 of the preamble to todays final rule.
Percentage of Facilities Covered
However, EPA disagrees with the commenters characterization of EPAs assumption regarding
percentage of facilities covered. Please see EPA-HQ-OAR-2009-0923-1151-108 for EPAs
response to APIs conclusion that the onshore petroleum and natural gas production facilities
covered by the rule would exceed 4 percent of facilities nationwide.. In addition, the Economic
Impact Analysis did not assume any reporter would be permanently excluded from the
reporting program. Rather, EPA used best available information to estimate the number of
108
See Mandatory Reporting of Greenhouse Gases, 74 Fed. Reg. 56,260, 56,319 (Oct. 30, 2009).
109
110
See Mandatory Reporting of Greenhouse Gases, 74 Fed. Reg. at 56,363.
111

1005
facilities likely to meet the reporting threshold, and then used that information to estimate
monitoring and reporting costs. EPA expects entities that did not meet the reporting threshold to
incur the determination costs discussed in Section 4 of the Economic Impact Analysis (EIA) for
todays final rule found in docket (EPA-HQ-OAR-2009-0923).
Compliance Cost Estimate
EPA also disagrees with the comment that the cost estimated by the Agency is unrealistically
low for the compression sector. First, as described in preamble II.F, EPA has modified todays
final rule so that it does not require reporters to install meters that would necessitate equipment
shutdowns. This modification reduces burden while maintaining data quality. Second, as
described in preamble II.D, EPA clarified todays final rule to allow reporters to conduct an
annual measurement of each compressor in the mode as it exists at the time the annual
measurement is taken, with shutdown depressurized mode monitored for each compressor at
least once every three each years. EPA has also revised todays final rule for onshore
production reciprocating and centrifugal compressors to use emission factors for calculating
emissions, rather than an annual measurement. Please see Compressor Modes and Threshold
in EPA-HQ-OAR-2009-0923.
Furthermore, EPA is unable to evaluate the commenters compliance cost estimate of $14,000
per facility because insufficient documentation was provided to explain how this figure was
calculated. For example, the commenter did not provide assumptions for the labor rate(s), labor
hours, or a detailed labor description used to calculate the estimated burden. Regarding the
American Petroleum Institute (API) cost estimates referenced by the commenter, EPA has
provided detailed responses to API explaining why their estimates are too high; see EPA-HQ
OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923-1151-107 for further details.
EPA-HQ-OAR-2009-0923-1151-107 identifies source categories discussed in each comment.
EPA estimates the compliance costs per facility in the first year and subsequent years as $27,084
and $7,088 respectively. EPAs estimates include costs for registration; regulation review;
monitoring plan development; equipment; training; monitoring & measurement; documentation;
reporting; auditing; and archiving.
Regarding the comment about the burden for making a reporting determination, EPA concluded
that screening tools would facilitate reporting determinations and plans to make such tools
available. Accordingly, EPA has updated the Economic Impact Analysis to better account for
reporting determinations and expected use of screening tools; see EPAs complete response in
Section III.B.2 of the preamble to todays final rule. EPA estimated determination costs for nonreporters well below the threshold to use screening tools as well as the higher burden for those
entities that are close to, but do not exceed the reporting threshold, that will supplement the
screening tools with preliminary monitoring. Please see Section 4.4 of the Economic Impact
Analysis for a detailed discussion of these costs.
In addition, todays final rule allows reporters to request the use of best available monitoring
methods under certain conditions, such as lack of service providers and qualified technicians;
please see the response to EPA-HQ-OAR-2009-0923-1011-27 for more information.
1006
Incorrect NAICS Codes

Finally, EPA agrees with the comment that the proposed analysis listed incorrect NAICS codes
for processing and liquefied natural gas and has since corrected the analysis to include
processing under NAICS code 211 and LNG under 48620. However, EPA disagrees that this
error resulted in incorrect estimates of the cost impacts or that it underestimated the costs. EPA
developed the costs per facility per segment independently and prior to allocating segments to
NAICS codes. That is, EPA aggregated the costs by NAICS code after the costs were estimated
within each segment on a facility-basis. The results of the screening analysis to determine
whether todays final rule would have a significant impact on a substantial number of small
entities also show that the NAICS error did not cause an incorrect conclusion. The analysis for
todays final rule, which assigned costs to the correct NAICS codes, resulted in the same
conclusion as the proposalthe rule will not have a significant economic impact on a substantial
number of small entities. See Section 5 of the Economic Impact Analysis for complete details
and results.

EPA has significantly underestimated the burden to industry to obtain the information to
calculate the emission estimate for gathering pipelines. It is also noted that the emission
factor for the gathering pipeline segments (2.81scf/hour/miles) is derived from an
unexplained total pipeline emission estimate of 6.6 Bscf. See TSD at 147-148. Anadarko
suggests that if EPA has already established this emission estimate, they should simply add it
to the rolled-up GHG inventory for the natural gas industry rather than requiring operators to
expend significant resources re-creating the exact same number.
Response: EPA disagrees with the commenter on the burden associated with estimating
emissions from gathering pipelines; the April 2010 proposal required reporters to so a simple
calculation (multiply the number of miles of gathering pipelines by the emissions factor provided
in the proposal). However, these costs are not relevant to todays final rule because EPA does not
require the reporting of emissions from gathering pipelines and boosting stations. See Section
II.F of the preamble for discussion about why todays final rule does not include gathering
pipelines and boosting stations.

The rule would be far more costly than EPA predicts. Anadarko has worked in
conjunction with the API and GPA to develop an estimate of the cost of compliance with the
1007
proposed rule for the upstream and midstream sectors. Anadarko concurs with and
incorporates by reference the cost estimates provided in API and GPAs comments to the
proposed Subpart W.
API analysis shows that total costs for the upstream sector will range from $1.8 to 3.5 billion
per year, as compared with EPAs estimate of about $28 million per year for the upstream
sector. APIs analysis shows that EPAs estimate is over two orders of magnitude low.
Further, APIs analysis only concentrated on the upstream sector and did not include
midstream sector costs.
GPA evaluated the economic impact for the midstream sector. That analysis shows that total
costs for all gas plants would exceed $698 million while the total costs to gathering/booster
stations in the midstream sector would exceed $3.8 billion for the initial reporting year.
It is clear that EPA has failed to recognize the full impact of requiring reporting at all
facilities and pipelines in the upstream and midstream sectors. EPA should delay the final
rule pending a re-evaluation of the Economic Impact Analysis and rationalization of the rule
requirements, rule approaches, burden, and costs versus the amount of emissions covered.
These analyses show that EPA can and should significantly simplify and streamline reporting
requirements for the upstream and midstream sector, and we urge EPA to do so for all the
reasons set forth in these comments.
Response: Regarding the American Petroleum Institute (API) cost estimates referenced by the
commenter, EPA has provided detailed responses to API explaining why their estimates are too
high; see EPA-HQ-OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923-1151-107 for
further details. EPA-HQ-OAR-2009-0923-1151-107 identifies source categories discussed in
each comment.
Regarding the commenters estimates for the midstream sector and the GPA cost estimates,
please refer to EPA-HQ-OAR-2009-0923-1151-107 and EPA-HQ-OAR-2009-0923-1206-25
respectively. EPA has assessed the commenters concerns about the processing segment and has
1) clarified specific requirements for reporters; 2) removed the requirement to report emissions
from gathering lines and booster compressor stations and 3) modified methodologies in a number
of areas, such as the monitoring of compressors in all the three modes of operations; please see
response to EPA-HQ-OAR-2009-0923-0055-16 and Section II.F of the preamble to todays final
rule for further details.
The rule would be far more costly than EPA predicts.
Member companies of AXPC have worked in conjunction with the API to develop an estimate of
the cost of compliance with the proposed rule for the upstream sector. AXPC concurs with and
incorporates by reference the cost estimates provided in API's comments to the proposed Subpart
W. This analysis shows that total costs for the upstream sector will range from $1.8 to 3.5 billion
per year, as compared with EPA's estimate of about $28 million per year for the upstream sector.
1008
API's analysis shows that EPA's estimate is over two orders of magnitude low.
It is clear that EPA has failed to recognize the full impact of requiring reporting at all facilities in
the onshore petroleum and natural gas production sector. API's analysis shows that EPA has
significantly understated the cost and burden of Subpart W as they are significantly higher than
EPA projected. AXPC requests that EPA re-evaluate the costs associated with this rule, and
should significantly simplify and streamline reporting requirements for this sector.
Response: EPA disagrees with the comment. Regarding the American Petroleum Institute (API)
cost estimates referenced by the commenter, EPA has analyzed those cost estimates and
determined that they are too high. A detailed analysis is available in EPAs responses to
comments EPA-HQ-OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923-1151-107.
These references also discuss changes made to the rule to reduce burden; see Section II.E of the
preamble to todays final rule for a comprehensive summary of the changes made to todays final
rule. Finally, EPA reviewed its Economic Impact Analysis (EPA-HQ-OAR-2009-0923) and
updated it to reflect changes in todays final rule.
Commenter:
In the Preamble (page 18610 of the Federal Register), the EPA states This proposed
supplemental rule incorporates a number of changes, including, but not limited to, different
methodologies that provide improved emissions coverage at a lower cost burden to facilities than
would have been covered under the initial proposed rule, the inclusion of onshore production and
distribution facilities While the original proposed rule called for reporting of 24 emission source
types, and this rule requires only 21, the EPA has lumped different types of pneumatic pumps
into more general categories, removed wet seal degassing and lumped fugitives into one general
category thereby resulting in a reduction in the number of reportable sources. Also, we concede
that the original proposed rule required direct measurement of fugitives and other sources using
HIVOL sampling, bagging, etc. However, the newly proposed rule includes reporting
requirements for many smaller sources that have limited or no emission data that were not
previously addressed as well as wellhead activities by basin and pipeline/gathering fugitives.
EPA is still requiring direct measurement of many of these sources, which is costly and time
consuming, and remains significant for operators to comply.
Response: EPA disagrees with this comment; please see EPAs response to EPA-HQ-OAR
2009-0923-1015-1 for discussion about EPAs basis for determining which sources to include in
the rule as well as revisions to todays final rule that reduce monitoring burden..

1009
Section 98.233(n): Flare Stacks: For flaring associated with well testing and the potential for
flaring emissions from dehydrators, unconventional well workovers and completions, and tanks,
the rule points to the calculation methods for flare stacks.
* For flaring associated with well testing, API assumed each gas well covered by the rule is
tested one time per quarter. The vast majority of well tests are conducted while the wells are in
operation and do not require flaring. As a result, API assumed 0.1% of the estimated 1,500,000
annual well tests were flared. Based on 1 hour to determine the GOR for each well, and $100 per
hour, results in a projected cost of $150K.
* No additional costs were assigned to flaring associated with dehydrators, unconventional well
workovers and completions, and tanks. For each of these sources, costs were estimated for
venting emissions from these sources (discussed previously). The flare equations would apply
the same volume estimates. Data management, which is addressed separately for all of the OPGP
sources, is the only cost that would apply to flared sources.
Response: The commenters estimated cost per well ($100) for testing and flaring is slightly
higher than EPAs estimate per well (approximately $71). The difference in costs per well results
from different assumptions about labor rates. EPA assumes one hour for a junior engineer to
perform an emissions calculation with GOR measurement at a $71 per hour labor rate. Please
refer to Section 4 of the Economic Impact Analysis (EPA-HQ-OAR-2009-0923) for how EPA
determined labor rates for the oil and gas industry. EPA cannot respond further to the
commenters labor rate assumption because they did not provide any information to substantiate
it.
Regarding assumptions about activity data, the commenter assumed a lower number of annual
well testing and flaring events (1,500) than what EPA assumed (3,000). EPA based its estimate
of 3,000 on the assumption that all wells drilled and completed would flare the gas produced
during testing. While EPAs activity data assumption is higher than the commenters, the
difference in national costs under each assumption is minimalroughly $200,000 assuming
3,000 flarings versus about $100,000 assuming 1,500 flarings.
Todays final rule allows reporters to use available GOR data from the wells being tested. EPA
assumes reporters will know this GOR data even in the absence of Subpart W, e.g., to determine
well flow rates and fluid characteristics as part of normal operations.
Section 98.233(o): Centrifugal compressor wet seal degassing vents: API did not evaluate this
source category since centrifugal compressors are not often used in production operations.
Section 98.232(c)(16) and Section 98.233(w): EOR injection pump blow-down API did not
evaluate this source category.
Section 98.232(c)(17) and Section 98.233(d): Acid gas removal (AGR) vent stacks API did not
evaluate this source category
1010
Response: EPA appreciates this comment.

Combustion Units non portable Presuming that the natural gas factors in Table C-1 of the
rule can be used throughout the Onshore Production Sector, which enables Tier 1 or Tier 2,
operators would be required to track the run time and load (company data) for each individual
combustion unit in order to calculate fuel volume and hence emissions. If Table C-1 factors
cannot be used, then the industry would be faced with Tier 3 metering which would impose
astronomical time and cost burdens which are not evaluated further. Partially based on EPAs
2008 Inventory of U.S. Greenhouse Gas Emissions and Sinks Table A-118, API estimates a
population of some 526,627 individual combustion units subject to the rule in the OPGP Sector.
These are comprised of the following: (1) Fired Separators = 307,734; (2) Other Heaters =
141,736; (3) Dehydrator Regeneration Heaters = 33,907; (4) Compressor Engines = 34,047; and
(5) Other Engines = 10,203. Despite the typically small size of the heaters (~0.3 MMBTU
average) and engines in the OGPG sector, the same company data will need to be tracked,
QA/QCd, and used to generate fuel volume use and hence emission estimates, as required for
combustion units at discrete facilities meeting the 25,000 metric ton threshold in the original
rule. Assuming half hour per combustion unit to track and collect company data, perform
adequate QA/QC, and calculate emissions at a rate of $100 per hour the total cost is estimated as
$26.3 MM for this source category.
Response: The commenter has raised two issues about combustion emissions: use of the
appropriate method to calculate CO2 emissions (i.e., the Tier methodologies specified in Subpart
C) and the associated burden to estimate fuel volumes. Regarding the use of appropriate method,
EPA did not intend for Subpart W to require reporters to use either Tier 3 or Tier 4 to report
combustion emissions from onshore production. Therefore, EPA has clarified todays final rule
to require onshore production reporters to use either Tier 1 or 40 CFR 98.233(z), which is a
method in Subpart W similar to Tier 1, to estimate and report their combustion emissions. EPA
expects that onshore production reporters who combust fuels that are not listed in Table C-1
(e.g., field gas), and therefore not eligible for Tier 1, to use the method in Section 98.233(z) to
estimate and report emissions from combustion sources.
EPA agrees that the proposed requirement to monitor all combustion equipment in onshore
production would be burdensome. Therefore, the major combustion equipment subject to
reporting under todays final rule includes compressor/ generator engines and drilling rigs.
Todays final rule does not require reporting of emissions from external combustion units that
have heat capacity of 5 MMBtu per hour or lower. Instead, reporters will provide only an
activity count for external combustion equipment (fired separators, heaters, and dehydrators) at
or below 5 MMBtu per hour. EPA notes that the cost estimate for Subpart W process emissions,
specifically estimating major equipment leaks, assumes reporters do an activity count and
therefore also captures the burden for external combustion units to do these activity counts.
Furthermore, EPA has determined that nationally almost all of the heater/treaters, fired
1011
separators, and dehydrators fall below the 5 MMBtu per hour equipment threshold; hence there
is no burden to report from these sources.
EPA has determined that the combustion cost estimates for the April 2010 proposal, which are
used in todays final rule as well, account for the monitoring and reporting burden associated
with onshore production equipment. The April 2010 combustion cost estimate assumed that
reporters would use the Tier 1 methodology or its equivalent and therefore incorporated labor
assumptions that were not specific to a particular industry segment. In response to the comments
and final rule changes, EPA used assumptions tailored to the onshore production segment to
estimate the costs for them to determine fuel consumption at compressor engines and drilling
rigs. Given that compressor engine fuel consumption can be determined using heat rate and
operating time of the equipment, EPA assumed that an operator would spend 10 minutes
collecting these data for each unit. For drilling rigs, operators can collect fuel consumption
information from their contractors, who typically drill and complete the wells and therefore
know how much fuel the equipment consumes. EPA has therefore assumed 30 minutes will be
sufficient to determine the fuel consumption for each drilling event. The combustion estimates
resulting from these tailored assumptions are lower than the estimates relying on assumptions
relevant to a broader array of industry segments; see the memo, Additional Analysis of
Combustion Costs for Onshore Production, for the detailed analysis and results (EPA-HQ
OAR-2009-0923). EPA has decided, however, to retain the higher estimate in order to provide a
more conservative estimate of combustion-related costs. See Section III.A of the preamble to
todays final rule for further discussion about the estimation of combustion costs.

Section 98.233(z): Portable equipment combustion emissions - Presuming that EPA requires
reporting of portable combustion units in the final rule (issues associated with these requirements
are outlined in Section III, comments 7, 12, and 41), the use of Tier 1 methodologies should be
allowed for this source category. APIs analysis estimates some 317,926 potential portable
combustion units subject to the rule in the OPGP Sector. Although only a subset of these would
be located at an individual well-site (well head) for more than 30 days in a year, operators would
still have to track each instance and gather company records for each portable combustion unit
from their owners to later determine if they had exceeded the 30 day time limit and were subject
to reporting. Estimating 2 hours total to track each portable combustion unit instance, obtain fuel
and BTU input information from the owner, obtain run hours and load or fuel use from the
owner, QA/QC the information, and calculate fuel use and emissions for each unit at a rate of
$100 per hour yields a total cost of $31.8 MM for this source category.
Response: First, EPA has modified todays final rule such that reporters should not need to track
equipment operating at non-reporter facilities. Specifically, EPA has removed the 30-day time
limit for portable combustion units; reporters will provide information about equipment
1012
regardless of the duration of time each unit stays in one place. See response EPA-HQ-OAR
2009-0923-1170-7 for further details. Second, EPA disagrees with the labor assumptions and
costs presented in the comment. The commenter has applied the labor hours, including fuel use
tracking, BTU content, etc., to the entire population of portable combustion units (371,926) and
not the subset actually subject to reporting, which results in the unreasonably high $31.8 million
burden. See EPAs response to EPA-HQ-OAR-2009-0923-1151-104 for details about on EPA
revisions to combustion emissions reporting from onshore production and labor assumptions.
Note that the requirements for portable equipment are the same as those for non-portable
equipment.
Data Management System Set-up and Data Entry: To adequately manage the massive amounts of
data required by the rule and enable QA/QC, emissions calculations, reporting and
recordkeeping, each covered entity will have to set-up some type of data system. A cost range
was examined, with a low range cost of $100,000 per data system for a reporting company that is
adding the calculation elements for the rule into an existing system, and a high cost range of
$850,000 for a reporting company that has to implement a data management system to handle the
required information. Using EPAs estimate of 1,232 covered entities, this yields a cost range of
$123.2 MM to $1.05 B. Entering the required date into these systems is estimated to cost an
additional $47 MM for a total cost range of $169 MM to $1.1 B.
Response: EPA disagrees with the comments on both the data reporting software needs as well
as the associated costs. As discussed in Section III of the preamble to todays final rule, EPA
does not require the use of any data management software in todays final rule; it is at the
discretion of reporters to buy one for their own convenience. Accordingly, EPAs analysis
focuses on the activities and equipment required to fulfill todays final rules requirements.
Section III of the preamble to todays final rule summarizes the basis for EPAs disagreement
with this comment. The remainder of this response provides additional details underlying EPAs
analysis of this comment.
First, the estimate of 1,232 reporters cited by the commenter represents a combination of
operators and basins. EPA has estimated that under todays final rule that there will be 981
reporters for onshore production, which represents 557 unique operators. Approximately 92
percent of these 557 unique operators have a limited presence in one to three basins, as follows:
428 operators will report from one basin only; 59 from two basins; and 25 from three basins.
The remaining 8 percent (45 operators) operate in more than three basins. Although the operators
with a presence in more than three basins may have greater data management needs, they can
effectively use spreadsheet and database software because the monitoring methods require a one
time data collection, not a continuous data collection that might warrant complex data
management software. EPA has therefore concluded that operators subject to the rule do not
need sophisticated or customized data management software to monitor and report emissions
1013
from their limited operation size; they can effectively do that using spreadsheet and database
software in common use (such as MS Excel and MS Access).
Although not required under the rule nor viewed necessary by EPA, reporters may use
customized software packages for data collection. Reporters, for example, may use the
SANGEA software, which has been available for over a decade and free of charge, to
accommodate the rule data collection requirements. EPA has found that major oil and gas
operators already use software programs like SANGEA that can be adapted to accommodate
the data collection requirements for this rule.
Also, consideration of the unique number of reporters is key because those who choose to
develop or purchase a data collection system will likely obtain one for all of the basins in which
they operate, not one for each of the basins for which they report.
In sum, EPA determined that the commenters cost estimate is much higher than the data
management software that may be considered by reporters. Although the commenter did not
provide any information about the software represented in its analysis (except for cost), a system
in the price range assumed by the commenter is usually customized to accommodate data needs
that extend far beyond the scope of this rule. See Section III.B of the preamble to todays final
rule for further discussion. EPA concluded that the Economic Impact Analysis presents a
realistic estimate of the costs to collect and manage data under todays final rule.
API Cost Estimate Summary: As the following table illustrates, APIs projected burden and costs
for the Onshore Petroleum and Natural Gas Production sector under the proposed rule exceed
$1.83 billion and are larger than EPAs projection of $27.7 MM by a factor of at least 66 times.
APIs alternative cost estimate is not inclusive of all sources (e.g., acid gas removal) and applies
conservatively low estimates for many cost elements, such as meter costs, analytical costs and
labor time. Even with these conservative assumptions, the cost per metric ton of CO2 equivalent
emissions from APIs analysis exceeds $8.17 vs. EPAs estimated $0.18 per metric ton in Table
W-5 of the preamble.
1014
1015
Response: Except as noted below, EPA disagrees with this comment and has determined that
the commenters cost estimates are too high. See EPAs responses to the following comments
for detailed explanation of EPAs analysis and conclusions:
EPA-HQ-OAR-2009-0923-1151-89: for response to commenters estimates of the
number of wells and total emissions covered; the costs to inventory well and non-well
sites; costs for well bore configuration; assumptions about travel to the sites; and the
costs to make reporting determination
EPA-HQ-OAR-2009-0923-1151-90: for response to the commenters cost estimate for
dehydrator units
EPA-HQ-OAR-2009-0923-1151-91: for a response to the commenters cost estimate for
tanks in hydrocarbon liquids service
hydrocarbon tanks and the hydrocarbon liquids dissolved CO2 category
EPA-HQ-OAR-2009-0923-1151-93: for a response to the commenters cost estimate to
sample hydrocarbon liquids dissolved CO2 and produced water dissolved CO2
reciprocating compressor rod packing venting
pneumatic pumps
well venting for liquids
unconventional well completions and workovers
conventional well completions and workovers
1016

associated gas venting and flaring
well testing flaring
EPA-HQ-OAR-2009-0923-1151-101: for response to the comment about centrifugal
compressor wet seal degassing vents
EPA-HQ-OAR-2009-0923-1151-102: for response to the comment about EOR injection
pump blowdown
EPA-HQ-OAR-2009-0923-1151-103: for response to the comment about acid gas
removal vent
non-portable combustion units
portable combustion units
data management
EPA-HQ-OAR-2009-0923-1151-129: for response to commenters cost estimate for
produced water emissions
EPA-HQ-OAR-2009-0923-1151-5: for details regarding EPAs changes to address other
commenters burden concerns.
Estimated Burden on Oil and Gas Production
The expansion of the MRR to include the entire natural gas supply chain, from dispersed wells
and small facilities to distribution meter stations and city gates; tremendously expands the scope
of the rule in terms of number of sites, source types, and monitoring/calculation methodologies.
Given the size and complexity of the U.S. oil and gas production, gathering, processing,
transmission, storage, and distribution sector; with over 823,000 producing wells, tens of
thousands of ancillary facilities, thousands of miles of collection, transmission, and distribution
pipeline systems, millions of individual pieces of equipment and components, and complex
operating and ownership arrangements; it is imperative that a simple and innovative system be
defined that balances the data collection and reporting burden against the amount of GHG
emissions quantified.
API believes that EPA has significantly understated the cost and burden of the rule in their
Economic Impact Analysis for the newly included Onshore Petroleum and Gas Production
(OPGP) category. APIs projected burden and costs for the Onshore Petroleum and Natural Gas
Production sector under the proposed rule exceed $1.8 billion, significantly larger than EPAs
projection of $27.7 MM. The cost per metric ton of CO2 equivalent emissions from APIs
analysis is $8.17 compared to EPAs estimated $0.18 per metric ton as stated in Table W-5 of the
preamble. Details on how we reach this conclusion are provided in a separate cost analysis
1017
described in Section VI.

As illustrated in APIs cost analysis, the burden of the rule is significantly higher than EPA has
projected. Our estimates also show that in total over 2,200 man-years (based on 2,000 hours per
year) are required by the oil and gas production sector in the first year just to meet the
compliance requirements. As a result, we firmly believe that EPAs timing goals are
unachievable given the vast number of sites and sources subject to this reporting rule that have
not previously had regulatory reporting or inventory requirements, coupled with the
extraordinary amount of work that must be done to implement the rule. To illustrate the
impossibility of immediate implementation: if the average well site takes 2 hours to drive to and
inventory components, it would take 667 man years (based on 2,000 hours per year) just to
inventory the fugitive components and equipment on the 667,000 wells covered under the rule.
Absent withholding the Onshore Petroleum and Natural Gas Production portion of the rule, API
recommends simplifying several portions of the rule, and a phased rule implementation approach
in place of the current unrealistic deadlines.
To further streamline the burden imposed by Subpart W, EPA must provide a simplified
screening mechanism for onshore oil and natural gas operations to determine if they are required
to report. Without a threshold screening mechanism for any of the sector categories covered by
Subpart W, all oil and natural gas operations have to compile a GHG inventory to determine if
they exceed the threshold. EPA neglected to include the cost associated with this activity. In the
first year, 100% of the source categories will need to be estimated using the proposed methods to
determine the rule applicability.
API Requests:
To address this widely disparate projection of burden and costs, API requests the following:
- EPA should delay inclusion of the Onshore Petroleum and Natural Gas Production sector in the
final rule pending a re-evaluation of the Economic Impact Analysis and rationalization of the
rule requirements, rule approaches, burden, and costs versus the amount of emissions covered.
- To address the unprecedented amount of work and resources necessary to implement the rule as
proposed, API requests the use of BAMM for source types requiring metered flow rates or
monitored parameters for the initial year. BAMM is particularly important for a sector like oil
and gas. Unlike a chemical plant or a refinery, oil and gas operations are spread out over a basin
and are often not staffed at all times. Often there is little existing infrastructure at these disparate
locations. Therefore, the use of appropriate meters and the counting of pieces of equipments will
be more difficult for these spread out oil and gas sources. In addition, given that many of the
sources are relatively minor individually and can be grouped together for estimation purposes,
API has proposed several simplifications that do not compromise data quality. Even with event
the Sub-basin entity simplifications that API has proposed, a phased implementation of the rule
over a period of several years will be necessary to fully understand the requirement of the rule,
inventory all sources subject to the rule, and develop data collection systems, etc. to comply with
the rule.
- When re-structuring and finalizing the rule, EPA should take all opportunities to simplify the
requirements and reduce burden. The rule should allow for simplified methods or outright
1018
exemptions for devices and operations that are below a size or threshold level. Such units may
include, but are not limited to, no-bleed pneumatic controllers; storage tanks and gas
dehydrators with low throughputs; small compressors, small combustion units and similar
sources.
- EPA must provide a simple screening approach to determine applicability. For example, a
Basin entity with 50 wells or less, at EPAs estimated emissions of 370 tonnes CO2e per well,
could be exempted from reporting.
Response:
Burden Comments: Monitoring Requirements and Scope of the Rule
EPA has carefully considered the complexity of the onshore production operational structure. As
discussed in todays final rule, EPA has decided to require reporters to monitor and report
emissions from specific sources using monitoring methods tailored to minimize the reporting
burden while maintaining the collection of reasonable-quality emissions data. For example, EPA
has revised todays final rule to require operators to count their major equipment instead of each
individual components. Other changes include allowing the use of existing available produced
gas composition data reducing the need to sample, setting equipment thresholds that require only
a subset of equipment to be monitored more rigorously, and allowing the use of engineering
estimation to determine equipment throughputs. See Section II.E of the preamble for all the
major changes to the rule.
Regarding the comments about balancing the compliance burden against the data collected, see
EPAs response to EPA-HQ-OAR-2009-0923-1014-5, which discusses the analyses EPA
conducted to consider these tradeoffs.
Compliance Cost Estimates and Underlying Labor Assumptions
EPA has also reviewed the commenters statements about the costs and weighed that input
against its Economic Impact Analysis. EPA has concluded that while some adjustments were
necessary (e.g. accounting for threshold determinations made by entities that do not meet the
emissions threshold), the Agencys methodology and assumptions were sound and relied on best
available data to estimate the costs. Please see Sections 4 and 5 of the Economic Impact
Analysis (EPA-HQ-OAR-2009-0923) for the changes to EPAs cost estimate since the April
2010 proposal. EPA disagrees with the commenters cost estimates. EPA has analyzed the
commenters cost estimates and determined that they are too high. A detailed analysis is
available in EPAs responses to comments EPA-HQ-OAR-2009-0923-1151-89 through EPA
HQ-OAR-2009-0923-1151-107. EPA-HQ-OAR-2009-0923-1151-107 identifies source
categories discussed in each comment.
In addition, EPA disagrees with the commenters estimate of the number of man-hours required
to report under the rule. Specifically, EPA disagrees with the commenters assumption that
reporters would have no reason other than data collection for Subpart W to visit each of the
wells. Please refer to EPA-HQ-OAR-2009-0923-1060-1 for further discussion.
Reporting Determination: Screening Tools and Cost Estimates
Regarding the costs for facilities to make a reporting determination, EPA agrees that the
Economic Impact Analysis would better reflect the rules total economic burden by including all
reporting determination costs. The Agency also agrees that screening tools would facilitate
reporting determinations and plans to make such tools available. Accordingly, EPA has updated
1019
the Economic Impact Analysis to better account for reporting determinations and expected use of
screening tools; see EPAs complete response in Section III.B.2 of the Preamble to todays final
rule.
Recommendations for BAMM and Timing of Rule Implementation
Regarding the commenters recommendation to delay implementation of the rule, see Section
II.F of the preamble a complete response and discussion about conditions under which reporters
may use BAMM. Regarding the phased-in implementation, EPA has modified the emissions
monitoring requirements such that reporters can gather sufficient data in the first reporting year
itself. Hence, EPA does not see the need to conduct a phased-in implementation. For any
exceptional circumstances under which data cannot be gathered in the first year, the reporters
have the option to request BAMM.
API believes that EPA has significantly understated the cost and burden of the rule in their
Economic Impact Analysis particularly for the newly included Onshore Petroleum and Gas
Production (OPGP) category. Our analysis, which is not inclusive of all sources and costs,
applies conservative (i.e., low) values for many of the estimates. Even with these conservative
assumptions, our analysis suggests that EPAs projection of first year costs of $27.7 MM
(preamble Table W-5) for this category is too low by more than an order of magnitude. Details of
how we reached this conclusion follow:
- Although EPA notes that only 4% of the potential OPGP facilities (defined at the Basin level)
will be covered under the rule (preamble Table W-2) at the proposed 25,000 metric ton
threshold, EPA also determined that some 81% of the emissions from this category would be
covered. It is logical to assume that to achieve 81% emissions coverage, close to 81% of the sites
and equipment comprising the facility as defined in the rule would be covered by the rule.
Assuming 81% of equipment and activities in the OPGP sector are included in the rule and with
some 823,000 wells in the Energy Information Administrations (EIA) 2008 report, this implies
rule coverage of about 667,000 of these well sites and the requirement for inventory and
monitoring activities at each of them.
- Since the rule fails to provide a simple method of determining applicability, emission
estimation is required at each individual well site and associated surface facility in this category
just to determine if the reporting threshold is exceeded.
- Based on an analysis of wells per operator in each basin, the 667,000 count is likely low. (API
can provide EPA with additional information on this analysis.) Applying emissions per well
estimates of 5,000 (EPAs high range estimate), 700 (EPAs mid-range estimate), and 370
(EPAs low estimate) metric tons of CO2e per well yields coverage of 846,142 wells, 737,985
wells, and 668,898 wells respectively.
- At each of these sites, owners/operators will have to undertake a detailed inventory of
equipment and fugitive components on the site. As structured, the rule would require visiting
1020
each well site to determine at least the following:

* Section 98.232(c)(1) and Section 98.233(a): High Bleed Pneumatic controller inventory with
model number
* Section 98.232(c)(2) and Section 98.233(b): Low Bleed Pneumatic controller inventory
* Section 98.232(c)(3) and Section 98.233(c): Pneumatic Pump inventory by model number
* Sections 98.232(c)(10), (18), and (20) and Section Section 98.233(j), (x)n and (y): Tanks in
both water and hydrocarbon liquid service
* Section 98.232(c)(11) and Section 98.233(p): Reciprocating Compressors by model number
and horsepower
* Section 98.232(c)(21) and Section 98.233(r): Fugitive Component Inventory
* Subpart C: Combustion Unit Inventory
- Assuming that driving to each well site, inventorying the equipment and fugitive components
and logging them for future entry into a database takes an average of 2 hours and that the average
charge for a person and vehicle to conduct this inventory is $100/hr, the inventory of well sites
alone would suggest a cost of about $133.4 MM for the initial inventory of well sites (667,000)
which is 4.82 times EPAs total cost estimate for the OPGP sector.
- This same type of inventory activity would have to be conducted at other sites (i.e., non-well)
within the OPGP category. Assuming one such site for every 100 wells with 3 hours per site (due
to the larger nature of associated sites) and the same cost per hour, this would add an additional
$2.0 MM in cost.
- In addition to the physical site inventory, the well bore configuration (depth, tubing size, casing
size, packer arrangement, etc.) would have to be verified for each individual well to enable
estimation of blow-down emissions for both the well venting for deliquification source type. It
is assumed that this can be accomplished by a file review (electronic or hard copy) and will not
require any physical inventory. Assuming 0.5 hours per well and fully loaded cost of $100 per
hour for an engineering technician to review and verify the well-bore configuration adds an
additional $33.3 MM in cost.
Response: Regarding the comment that EPA underestimated the costs, see EPAs response to
EPA-HQ-OAR-2009-0923-1151-5 for a summary of EPAs assessment of the commenters
estimates and the Agencys estimates.
Percentage of Wells Reporting to Subpart W
In addition, EPA disagrees with the commenters assumption about the number of wells subject
to reporting under the rule, specifically, that 81 percent of the wells account for 81 percent of the
emissions. This assumption, which resulted in much more labor and complex monitoring than
required under the proposal, is incorrect. The magnitude of emissions from wells will vary
because wells produce different volumes of natural gas and petroleum. Therefore, monitoring 81
percent of the wells does not necessarily account for 81 percent of emissions.
Instead of assuming that percentage of the rules national emissions coverage (81) equals the
percentage of national wells covered (81), EPA used data on actual production volumes to
estimate the emissions per well. Specifically, EPA conducted the threshold analysis using actual
data available through the commercial database from HPDI LLC, which collects these data
1021
primarily from individual petroleum and natural gas producing States that require petroleum and
natural gas producing companies to report field data. The HPDI database includes operator well
count. In most cases, HPDI provides data for each well on the production of petroleum and
natural gas by operator and basin; some data are listed by property, which is a collection of
wells. EPA developed a reasonable estimate of the emissions per well by apportioning the
national emissions from each emissions source type to each of the wells based on the
contribution of petroleum and natural gas production from each well to the national total. This
analysis suggests that approximately 60 percent of the wells are covered by the reporters, not 80
percent.
Furthermore, 60 percent is a conservative estimate for the number of potential site visits because
multiple wells are typically connected to one set of equipment located at one of the wells. For
example, the U.S. National GHG Inventory estimates that there are 2 to 3 wells connected to one
separator. Hence, reporters will not need to visit all of their wells covered by the rule for
detailed data gathering; EPA estimates that approximately 30 percent of the well pads in the U.S.
will have equipment that may be visited for detailed data gathering.
In sum, the commenters assumption about the number of wells covered produced a higher
estimate of the number of wells covered, which in turn resulted in significantly higher estimates
of compliance costs.
Cost to Visit Well Sites
In addition, EPA disagrees with the commenters estimates of the costs to visit well sites and
non-well sites to collect data for compliance with Subpart W. EPA has accounted for costs to
gather rule-related data at the well sites assuming that reporters will accommodate data gathering
during their routine business-as-usual site visits. See response EPA-HQ-OAR-2009-0923-1060
EPA also disagrees with the commenters estimates for well blow-down calculations, in
particular the assumption that reporters would need to review and verify well-bore
configurations. Reporters will need four parameters to do well blow-down calculations: well
depth; well tubing diameter; well casing diameter; and well shut-in pressure. Oil and gas
operators already submit most or all this information to individual state oil and gas commissions
on an annual basis. In addition, this is basic data on wells that operators maintain internally.
Therefore, EPA determined that reporters will not need to spend additional time verifying data
about well-bore configurations that will have been recently verified for state or internal purposes.
Accordingly, EPA deems the commenters cost estimate as unnecessarily high.
Determination Cost and Screening Tools.
EPA agrees that screening tools would facilitate reporting determinations and plans to make such
tools available. Accordingly, EPA has updated the Economic Impact Analysis to better account
for reporting determinations and expected use of screening tools; see EPAs complete response
in Section III.B.2 of the preamble to todays final rule.
1022

Section 98.232(c)(14) and Section 98.233(e): As proposed, the rule requires an inventory of each
covered dehydrator unit for a variety of physical and operational parameters along with sampling
of the inlet feed gas for composition and water content and the outlet gas for water content.
Assuming 81% coverage of the 41,861 dehydrators in EPAs 2008 Inventory of Greenhouse Gas
Emissions and Sinks for the Natural Gas Production Stage yields 33,907 dehydrators covered
under the rule. An inventory of each of these is assumed to occur while inventorying well-sites
and assumed to add 0.5 hours of contractor time at $100/hr yielding $1.7 MM in costs.
Additionally, sampling and analysis of the dehydrator inlet and outlet gas for water content and
the inlet gas for composition is projected to cost $25 for each water content analysis and $475 for
each gas sample and extended compositional analysis. An additional 1.5 hours is projected for
collecting the samples, set-up and modeling of each dehydrator at a rate of $100/hr. Applying the
combined analytical costs of $525 per dehydrator to the 33,907 covered dehydrators yields an
additional $17.8 MM in costs. Combined with the labor costs, this yields a total of $24.6 MM for
dehydrator inventory.
Response: EPA disagrees with the commenters cost estimate for dehydrator units, in particular
the number of dehydrators subject to reporting. First, EPA disagrees with the assumption that 81
percent of the dehydrators listed in the 2008 Inventory of Greenhouse Gas Emissions and Sinks
for the Natural Gas Production Stage; see the response to EPA-HQ-OAR-2009-0923-1151-89 for
discussion about why it is inaccurate to assume 81 percent of sources are subject to monitoring.
Instead, EPA obtained its estimate that 24,405 dehydrators will be subject to reporting under the
rule from its threshold analysis, discussed in EPA-HQ-OAR-2009-0923-1151-89 and the T
0923).
Second, EPA disagrees with the commenters assumption that 0.5 hours are needed to collect
relevant information or that it should take 1.5 hours to run a software simulation run. Field
operators know their equipment well enough to provide the necessary inputs without much
deliberation. Furthermore, EPA has conducted GlyCalc runs and has determined that it takes
less than 10 minutes to run a simulation and copy the results into a data collection sheet.
Moreover, 10 minutes is a conservative estimate because it does not account for simulation
software configured to conduct multiple runs in tandem, which would not limit the user to
running one dehydrator simulation at a time. The simulation software can be easily configured
for multiple runs and would reduce the run time by up to 90 percent. Also, EPA assumed a junior
engineer at $71.03 per hour would be performing this work. Please see Section 4 of the
Economic Impact Analysis (EPA-HQ-OAR-2009-0923) for how EPA developed this labor
category and labor rates. EPA cannot respond to the commenters assumed labor rate ($100 per
hour) because they did not provide any other information.
Third, EPA never intended for reporters to take samples to determine inputs to the simulation
model. Therefore, EPA has clarified in todays final rule that reporters can use the default values
available in simulation software. Please see preamble Section II.F for more details on using
default software simulator composition values. Also, EPA has revised the requirements by
setting a dehydrator equipment threshold, eliminating need to simulate emission from the
1023
majority of the dehydrators. Hence, the dehydrator costs do not apply. Please see Section II.F of
the preamble for further details on the equipment threshold.
Section 98.232(c)(18) and Section 98.233(x): Closely related to tanks in hydrocarbon service is
the Hydrocarbon Liquids Dissolved CO2 category. Although EPAs Technical Support
Document discusses this category only in relation to CO2 EOR operations, the rule requires
quarterly sampling and analysis of the hydrocarbon liquid in all tanks in the production category
after they have equilibrated to ambient conditions. (Please note that the efficacy of this category
is discussed in another API comment specific to the source category.) Assuming that there are
0.25 tanks per oil well and between 0.2 and 0.92 tanks per gas well, yields a range of tanks
nationally from 183,000 to 514,000 in the production category. Applying the assumption that
81% of the tanks are covered by the rule results in an estimated number of tank ranging from
148,000 to 416,000. Based on this projected number, quarterly sampling, and a cost of $475 per
sample gathering and analysis, this yields a cost ranging from $271 MM to $791 MM for this
emission source.
Response: EPA did not intend to require quarterly sampling and analysis of the hydrocarbon
liquid in all tanks in the production category. Therefore, EPA has clarified in todays final rule
that only hydrocarbon liquids at EOR sites have to be sampled on an annual basis. See response
to comment EPA-HQ-OAR-2009-0923-1011-21 for further details. EPA has estimated that about
21 EOR operations will have to report to the rule. Hence the number of samples will be
equivalent to the number of tank batteries at these EOR operations. Therefore, todays final rule
requires much fewer samples for tanks in hydrocarbon service than assumed by the commenter.
Also, todays final rule does not require reporting of any produced water emissions. See response
to comment EPA-HQ-OAR-2009-0923-1151-129 for further details..
Section 98.232(c)(4) and Section 98.233(f): For well venting for liquids unloading the rule has
two options which will be detailed separately:
* Method #1 requires measurement of the vent rate during liquids unloading from each
reservoir/tubing size combination in each field. Examining EIAs 2008 field list, API estimates
15,716 (81%) of the 19,402 gas fields are included in the rule. Based on an assumption of one
producing reservoir per field, 3 discrete tubing sizes per field, and 2 operators per field, some
94,000 discrete measurements would have to occur. At a projected cost of $1,000 per
measurement, for temporary piping modifications, installation of a temporary recording flowmeter, conducting the measurements, and returning the piping to its original configuration, the
cost of measurement would be $94 MM. Assuming 2.5 hours per field/tubing size/operator
1024
combination (94,000) and $100 per hour to track venting by combination and perform the
calculations for emissions estimation an additional cost of $23.5 MM is projected. The total
projected cost for this category is $117.9 MM.
* Method #2 requires calculation of the blow-down volume from de-pressuring each individual
well-bore along with tracking, at an individual well level, the number and time of each individual
venting event. With some 373,000 gas wells subject to the rule (81%) and assuming EPAs
estimate that 41.3% of these wells vent for liquids unloading, this source will require about
154,000 individual well-bore de-pressuring volume calculations. Using an average of 31 discrete
vent instances per venting well per year (EPAs estimate) would require tracking of the number
and time of some 4.7 million discrete venting instances per year. To arrive at an emission
estimate would require subsequently calculating emissions for each of the 154,000 wells and
then summing these emissions, on a Basin entity level, for annual emissions. Assuming 0.5 hours
total per well to calculate the well-bore de-pressuring volume for each affected well; 0.1 hours
per event to track, QA/QC, and log the time; and 0.2 hours per well to calculate and QA/QC the
vent volumes to arrive at an emissions estimate, at $100 per hour this yields an estimated 293
man years of time (@2,000 hours/yr) and $58.5 MM cost for this method.
Response: EPA disagrees with the commenters assumptions and resulting cost estimates. EPA
proposed two options for reporters to monitor well unloading, described as Method 1 and
Method 2 by the commenter above. Method 2 is less costly than Method 1. Under the proposal,
reporters could use the costlier option but were not required to do so; same is true of todays final
rule. Therefore, EPAs analysis assumed reporters would use the lower cost option (Method 2) in
its analysis and did not evaluate the higher cost option.
EPA has analyzed the commenters estimate of the costs for Method 2 and has determined that it
is too high because it is based on four incorrect assumptions. First, the assumption that 81
percent of wells are subject to reporting leads to an overestimate of the number of wells requiring
blowdowns (154,000). Please see EPAs response to EPA-HQ-OAR-2009-0923-1151-89 for
why 81 percent of U.S. wells are reporting is too high. Instead, EPA has estimated that 137,662
wells requiring blowdowns will need to be reported, based on its assumption that 41.3 percent of
the reporting wells will need a blowdown. EPAs 41.3 percent assumption was derived from well
blowdown data found in GRI & EPA (June 1996) Methane Emissions from the Natural Gas
Industry. Volume 6. Second, EPA does not agree that it will take 0.5 hours to calculate emissions
per well. The calculation method provided by EPA can be set up in a spreadsheet or database
such that once the input data is available, the spreadsheet or database can replicate the
calculation for each well within a few minutes for an entire basin.
Third, EPA disagrees that an operator would need 0.1 hours per well to log the time for well
opening and shutting to atmosphere in a data sheet; most likely less than a minute per well would
suffice because the operator only needs to note the time the well was opened to the atmosphere,
the time the well was closed, and the duration of time that the well was open to the atmosphere.
Also, the 0.2 hours for calculating and QA/QC per well is redundant with the assumed increment
of 0.5 hours per well to calculate the well-bore de-pressuring volume. Finally, once a spreadsheet
or database is set up correctly, the electronic system replicates all of the calculations and requires
1025
minimal QA/QC. Therefore, the time assumed to set up the spreadsheet or database does not
apply to each calculation on a well-by-well level.
Considering all the points above, EPA has determined that its assumptions and methodology are
valid and result in a realistic cost estimate for reporting well blowdowns at about $733,000.
Section 98.232(c)(6) and Section 98.233(g): Unconventional Well Completions and Work-overs:
The proposed rule details two different methodologies for estimating emissions from
unconventional well clean-up, post fracture stimulation, on completion or work-over (which
appears to be defined as re-stimulation by repeat hydraulic fracturing in the rule) emissions
which are discussed separately. For both methods, the following assumptions were applied to
quantify the number of measurements:
* Completion: With an average of 28,800/yr wells drilled in the 2007 2009 period it is obvious
that drilling and completion did not occur in every field covered by the rule. Assuming 25% of
the estimated 15,716 impacted gas fields had active drilling and completion, that 90% of these
were fracture stimulated and thus unconventional, and that there were 2 covered operators per
field, yields an estimated range of 7,072 measurements that would be required.
* Work-over: Assuming that 50% of the existing covered gas wells are unconventional, that 10%
of these are re-stimulated annually (EPAs estimate), that between 1% and 25% of the fields
have active re-stimulation each year, and that there are two covered operators per field, yields an
estimated 157 to 3,929 work-over measurements per year.
* Method #1 requires the measurement of flow rate during clean-up for one well completion and
one well work-over in each field. From this a flow rate per minute is derived and applied to the
minutes of flow-back from each completion and work-over in a field. Applying an estimated cost
of $1,000 per measurement to between 7,200 and 11,000 estimated measurements yields cost
ranging from $7.2 MM and $11 MM for measurement. Assuming that tracking the vent time per
completion and work-over and performing the calculations requires 1 hour of time at $100 per
hour yields a cost of $3.9 MM which brings the total range to $11 MM to $15 MM for this
category/methodology.
* Method #2 requires monitoring across the completion choke and calculation of flow rate during
clean-up for one well completion and one well work-over in each field. From this a flow rate per
minute is derived and applied to the minutes of flow-back from each completion and work-over
in a field.
Applying an estimated cost of $500 per differential pressure measurement and subsequent
calculation to the estimated 7,200 to 11,000 measurements yields a cost range of $3.6 MM to
$5.5 MM for differential pressure tracking and flow modeling. Assuming that tracking the vent
time per completion and work-over and performing the calculations requires 1 hour of time at
$100 per hour yields a cost of $3.9 MM which brings the total cost range from $7.6 MM to $9.5
MM for this category/methodology.
Response: EPA disagrees with the commenter on the estimate of well completion and well
workover activity data used for burden analysis, as explained below. EPA determined that
1026
installing a flow meter on at least one completion per field, cited as Method #1 in the comment
above, is the lower cost option. EPAs analysis assumed reporters would use the lower cost
option (Method #1) and therefore did not evaluate the higher cost option, Method #2.
Well Completions
Instead of assuming 25 percent of the reporting fields have active completions and that 90
percent of these completions are hydraulically fractured, EPA used actual year 2006 data from
HPDI, LLC to determine which fields had any form of completions. The data showed only
1,177 unconventional field and operator combinations with any form of well completions in the
year 2006. Since the rule requires only one sample of well completion emissions per field, these
1,177 unconventional field combinations equal 1,177 samples. Of these samples, only 435 are
covered by reporting entities, far fewer than the 7,072 measurements estimated by the
commenter.
Workovers
The commenter incorrectly assumed that 50 percent of the gas wells covered by the rule are
unconventional. EPA estimated from the HPDI database that the actual number is 88,588 gas
wells or 20 percent. EPA cannot comment on the commenters assumption of 50 percent
because they did not substantiate it.
Based on the HPDI database, EPA estimates that there were a total of 1,535 operator and
unconventional gas well fields combinations in year 2006. Assuming each of these combinations
will have at least one sample workover measurement, EPA estimated that a total of 1,535
samples will be collected for year 2006, less than the 3,929 estimated by the commenter.
Cost estimates
EPA agrees with the commenters cost estimate per gas well sample, but disagrees with the total
estimates because they based them on inaccurate activity data. Also, EPA disagrees with the
commenters estimated labor rates; please see EPAs response to EPA-HQ-OAR-2009-0923
1151-90 for a response to this issue.
Section 98.232(c)(5) and Section 98.233(h): Conventional Well Completion and Work-over: For
conventional well completion and work-over, the rule specifies tracking of venting time per
individual well completion or work-over and then applying the wells production rate to calculate
the emission volume. Based on nearly 373,000 gas wells impacted by the reporting rule, and
assuming 50% are conventional and 10% are re-stimulated annually, results in an estimated
18,600 work-over procedures. An estimated 2,300 completions per year was based on 81% of the
number of gas wells drilled and an assumed 10% are conventional completions. Combined with
these values, an assumption of 1 hour of time at $100 per hour to track the vent time per
completion and work-over, and 1 hour of time at $100 per hour to perform the calculations, the
projected cost is $2.1 MM.
1027
Response: EPA disagrees with the commenters activity data assumptions and costs estimates.
EPA estimates of 1,663 fields with conventional completions and 17,931 conventional
workovers are based on actual data from HPDI database rather than an assumption. Also, EPA
does not see the need for 1 hour to track time taken for completion; this is a matter of logging
when the well completion started and when it ended, which would require, at most, a few
minutes. Please see Table 4-13 in the Economic Impact Analysis (EPA-HQ-OAR-2009-0923)
for the exact time estimated for a junior engineer and senior operator to determine the emissions
from conventional well completions and conventional well workovers. Finally, EPA does not
agree that it will take 1 hour to simply multiply two terms (completion/ workover time and well
flow rate) for each well. It does not take that amount of effort to do this calculation manually,
and even less time when using a spreadsheet or database. Hence, the commenters cost estimates
significantly overstate the burden.

Section 98.232(c)(13) and Section 98.233(m): Associated Gas Venting and Flaring: EIA tracks
the volume of natural gas vented and flared in the US. In 2006, EIA reported only 0.56% of
gross natural gas production in the US was vented or flared. As a result, the application of the
emission estimates for flares is not expected to impact many sources. For associated gas flaring,
API assumed that 0.5% of associated gas wells impacted by the rule flare gas, this results in 972
wells that would report flared emissions. Assuming 1 hour to determine the GOR for each well at
$100 per hour, results in a projected cost of $97K.
Response: In the analysis for todays final rule, EPA assumed that all of the Williston basin is
venting and/or flaring their associated gas, which involves approximately 4,421 wells. EPA
based this estimate on experience from the Natural Gas STAR program. The dataset cited by the
commenter is less preferable in this case because it includes emissions from sources other than
those covered by this requirement, such as well completions and workovers.
Furthermore, todays final rule allows reporters to use existing GOR data, which will minimize
the costs to report emissions from this source. See Sections II.C and II.F of the preamble for
more information about EPAs decision to allow use of existing GOR data instead of requiring
sampling.

Subpart W GHG reporting requirements include 21 emission sources for onshore petroleum and
natural gas production. In addition, affected facilities (reporting areas) are required to report
1028
emissions from Subpart C combustion sources and also have requirements for characterizing
field gas composition. These extensive reporting requirements - which require equipment
calibration, equipment surveys, measurements, process samples, and recordkeeping - place an
excessive and disproportionate burden on the onshore production sector. The locations of
onshore exploration and production (E&P) sources (i.e. large geographic distribution), and lack
of onsite power to automate data collection (e.g. manual data logs would be required)
significantly add to the resources needed to implement the data collection and reporting
requirements. The result is an excessive cost burden that has been significantly underestimated
by EPA.
Response: EPA does not agree with the comment. For a response to the comment about the
estimated costs for onshore production, please see EPAs response to EPA-HQ-OAR-2009
0923-1151-107. Please see Section II.F in the preamble for further details regarding changes to
field gas composition requirements. In addition, see EPAs response to EPA-HQ-OAR-2009
0923-1151-5 for discussion about travel to exploration and production sources; see EPAs
response to EPA-HQ-OAR-2009-0923-1151-104 and EPA-HQ-OAR-2009-0923-1151-105 for
discussion about the costs to report incremental combustion emissions. Finally, this rule requires
a one-time monitoring of emissions, not continuous monitoring. Therefore, EPA disagrees with
the commenters assumption that reporters will need automated data collection and onsite power
for this purpose.
Mandatory Reporting Rule (MRR) compliance for onshore petroleum and natural gas production
will require extensive effort including, but not limited to, direct emission and process
measurements (e.g. flow meters), thousands of quarterly and annual process samples (e.g.
storage tank liquids, separator liquids, gas samples), thousands of equipment surveys,
calculations and data management for thousands of emission sources, and project management,
record-keeping and reporting. Table 1 presents estimated Noble Energy costs for MRR
compliance including first year (Year 1) and subsequent year (Year 2+) estimated $/tonne CO2e
costs for each emission source and the entire Noble inventory. These cost estimates are based on
the emission measurement and estimation methods prescribed in Subpart W and Subpart C, and
GHG emission estimates from the Noble Energy 2008 inventory. Details regarding the data,
methods, and assumptions used for these cost estimates are provided in Attachment A. These
estimates provide guidance regarding emission sources where alternative, simpler emission
estimation requirements and methods are needed for reasonable compliance costs.
1029
1030
Attachment A Documentation of Estimated MRR Compliance Costs for Noble Energy

Onshore Petroleum and Natural Gas Production Operations
This attachment describes the data and methods that were used to estimate Mandatory Reporting
Rule compliance costs for Noble Energy onshore petroleum and natural gas production sources
affected by Subpart W and Subpart C. Compliance costs were estimated for most of the affected
emission sources and for preparing the entire inventory. The GHG emissions data that are the
basis for the majority of the estimates are from the Noble Energy 2008 GHG emission inventory.
This inventory is primarily based on emission estimation methods from and consistent with the
API Compendium. For some of the emission sources that are not included in the Noble
inventory, emissions were estimated using Subpart W or Subpart C methods if applicable activity
data were available. Emissions could not be estimated for some sources.
Table A.1 provides summary information for each affected emission source. As available, CO2e
emission estimates from the Noble 2008 are presented with estimated compliance costs for the
first year (i.e. Year 1) and subsequent years (i.e. Year 2+). Estimated compliance costs are
presented as $/tonne CO2e for the emission source. These costs include physical sample
collection and analysis and/or direct measurements, engineering calculations (e.g. E&P Tanks),
equipment calibrations, personnel training, recordkeeping, emission source level and rolled up
calculations, and reporting. Based on Nobles interpretation of the rule requirements, the last
column details the data, methods, and assumptions (e.g. survey and/or measurement of different
emission sources during a single site visit) used to estimate the costs. As would be expected at
this stage of a rulemaking, there is considerable uncertainty in these cost estimates. In addition
to the assumptions discussed in the Table A.1, other factors that could impact costs include a
shortage of service providers and trained personnel, and excessive demand (i.e. industry wide,
millions of emission sources would require survey, sampling, and/or measurement), and
complications with field measurements and process sampling (e.g. pressurized separator water
and oil samples).
Year 2+ costs for emission sources that require a survey (e.g. components, pneumatic devices),
are assumed to be 15% of the Year 1 costs based on a Noble average annual operations
acquisition and organic growth rate of 15%. Year 2+ costs for emission sources that require
annual or quarterly sampling and/or direct measurement are generally assumed to be 90% of
Year 1 costs to account for efficiency improvements.
Average sampling + analytical cost per sample are included as applicable. For process sample
collection and analysis (e.g. field gas or separator liquids composition), industry budgetary costs
typically range from $150 to $250 per sample. The costs presented in Table A.1 are generally
consistent with or lower than these guidelines suggesting they could be a low-biased estimate of
the true costs.
1031
1032
1033
Response: EPA has reviewed the commenters cost breakdown for each source and responded
below. While EPA has responded to the information provided, the commenter did not provide
all the necessary calculations for EPA to assess the cost per metric ton for each source.
Well Venting for Liquids Unloading EPA does not agree with the comment. First, Method 2 is
considerably simpler and cheaper for most operators than Method 1 because Method 2 is
predominantly a desktop calculation. The data necessary to do this calculationwell
tubing/casing diameters, shut-in pressure, and well depthare well-known to the operators
because most States require reporting of this data on an annual basis (see EPA-HQ-OAR-2009
0923-1151-96). The only new input required to do the Method 2 calculation is the time taken for
blowdown. Specifically, when the blowdown is manual, the operator conducting the blowdown
can keep a simple log and when the blowdown is based on a timer (e.g., in cases where there is a
plunger lift), it is simply a matter of gathering the timer data. None of these tasks are
burdensome. In sum, EPA disagrees with the commenters assumptions and high cost estimate
to monitor this source because the data are available and easily attainable.
1034
Associated Gas Venting and Flaring EPA has revised todays final rule to allow for use of
existing GOR data, thereby avoiding the need for reporters to make a GOR determination each
year. Most operators know their GOR, which will reduce the burden to monitor this source. See
also EPAs response to EPA-HQ-OAR-2009-0923-1151-99 for additional discussion about this
estimate.
Gas Well Venting during Unconventional Well Completions and Workovers EPA generally
agrees with the assumptions on time taken to monitor emissions. However, EPA does not agree
with the commenters methodology to calculate the cost. Reporters are required to only measure
one well completion and workover event per field. Please see Section II.D of the preamble for
further details. The commenter estimated the cost to measure all the well completions and
workovers events, 3,500, during a reporting period for 50 fields and consequently over-estimated
the cost. EPAs response to comment EPA-HQ-OAR-2009-0923-1151-97 provides additional
discussion about EPAs assumptions and estimates for this emissions source.
Gas-Fired Reciprocating IC Engines and External Combustion (Heaters, boilers) EPA has
significantly simplified the monitoring requirements for todays final rule. The reporters can use
Tier 1 from Subpart C or similar method from Subpart W using existing company records. This
will significantly reduce burden to report combustion emissions. See EPAs response to EPA
HQ-OAR-2009-0923-1151-104 and EPA-HQ-OAR-2009-0923-1151-105 for discussion about
the costs to report incremental combustion emissions.
Natural Gas Pneumatic Devices (low and high) and Pumps - EPA now requires the use of
emissions factors for all pneumatic devices in todays final rule. Hence, the reporters only need
to count the number of devices to estimate emissions. See Section II.E of the preamble for
further details.
Portable Combustion Sources (Drill Rigs) EPA notes that emissions from drilling rigs have to
be reported by the reporter, irrespective of operational control. See EPAs response to EPA-HQ
OAR-2009-0923-1151-104 and EPA-HQ-OAR-2009-0923-1151-105 for discussion about the
costs to report incremental combustion emissions.
Dehydrator (glycol) vent stacks Todays final rule does not require reporters to sample natural
gas in order to do the dehydrator emissions calculation. Instead, reporters may use default values
in the software program to estimate dehydrator emissions. See Section II.E of the preamble for
further details. EPA disagrees the commenters estimate of the time required to run GLYCalc.
See response EPA-HQ-OAR-2009-0923-1151-90 for EPAs complete response.
Components See response to EPA-HQ-OAR-2009-0923-1151-5 for discussion about changes
EPA made to todays final rule to reduce burden for onshore production, such as requiring
operators to count their major equipment instead of each individual component. Also, EPA does
not agree with the commenters assumptions about travel costs related to the monitoring
requirements. See response EPA-HQ-OAR-2009-0923-1151-5 for EPAs complete response.
1035
Produced Water Dissolved CO2 EPA has removed this source from todays final rule. See
EPAs responses to comments EPA-HQ-OAR-2009-0923-1151-129 and EPA-HQ-OAR-2009
Production Storage Tanks Todays final rule does not require sampling of low pressure
separator oil for tank emissions calculation and instead permits reporters to use default values in
software programs. See EPAs response to comment EPA-HQ-OAR-2009-0923-1151-90 for
more information about EPAs estimate of the costs for this activity. EPA does not have
sufficient details on the commenter analysis to respond in detail. For example, the commenter
did not specify their assumptions for the hours needed to run a tank emissions simulator or
collect the input data required.
Gathering Pipeline Fugitives - EPA has removed this source from todays final rule. See Section
II.E of the preamble for further details.
Reciprocating Compressor Rod Packing Vents In todays final rule, reporters can estimate
emissions from compressors in onshore petroleum and natural gas production using emissions
factor and activity count. This change from the proposal will significantly reduce burden to
reporters from 30 minutes to read a portable vane anemometer to a few minutes to count the
compressor and apply an emission factor. The commenters assumptions about the cost for
taking measurements are therefore no longer relevant.
Acid Gas Removal (AGR) Vent Stacks - Todays final rule allows reporters to use engineering
estimation under certain conditions to determine flow rate to an AGR. See EPAs response to
comment EPA-HQ-OAR-2009-0923-1024-26 for further details. This will significantly reduce
burden because it does not require reporters to install new meters. Therefore, the commenters
cost estimate for AGR is no longer relevant for todays final rule.
Hydrocarbon Liquids Dissolved CO2 In todays final rule, EPA has clarified that this
requirement applies only to EOR operations. Also, todays final rule has reduced the required
sampling frequency to an annual basis. These changes significantly reduce the cost estimate for
this source. Please see EPAs response to comment EPA-HQ-OAR-2009-0923-1011-21 for
further discussion.

In summary, Noble Energy has prepared a best estimate of the proposed rule costs based on its
understanding of the proposed rule requirements and experience developing GHG emission
inventories. However, rule changes and clarifications upon promulgation and as yet understood
external factors (e.g. limited service providers and excessive demand (i.e. industry wide, millions
of emission sources would require survey, sampling, and/or measurement), and complications
with field measurements and process sampling) could significantly increase the costs above these
1036
estimates. In addition rule implementation costs to develop data management and archival
systems will likely result in additional underestimated burden. Many of the proposed emission
estimation methodologies are cost-prohibitive and alternative simpler, streamlined methods need
to be provided. Alternative, simpler emission estimation methods are discussed in sub-section C.
Finally, as noted in Comment V, EPA should provide a practical applicability screening
approach for rapidly and efficiently determining Rule subjectivity based on the 25,000 tonne per
year CO2e reporting threshold. The inability to determine rule applicability with a reasonable
degree of certainty will require emission estimations for numerous small facilities to ensure
compliance certainty. This significantly adds to the regulatory burden and it does not appear
EPA has considered these costs for this rulemaking
Response: Please see EPAs response to EPA-HQ-OAR-2009-0923-1167-11 for a response to
the commenters best estimate of the costs. The Agency also agrees that screening tools would
facilitate reporting determinations and plans to make such tools available. Accordingly, EPA has
updated the Economic Impact Analysis to better account for reporting determinations and
expected use of screening tools; see EPAs complete response in Section III.B.2 of the Preamble
to todays final rule. Finally, see EPAs response to EPA-HQ-OAR-2009-0923-1151-106 for
information about the costs for data management and archiving.
The GHG emission estimation requirements in the Proposed Rule are overly burdensome and
EPA has not provided data quality objectives to justify the extensive costs. To address these
issues, Nobles recommends that the Proposed Rule revisions to reduce burden while collecting
GHG emissions data to develop a representative inventory that is no less useful include: removal
of insignificant emission sources; reducing the frequency of emission and process measurements;
adapting simpler, more cost-effective emission estimation methods; and representative sampling
of large source populations.
Nobles cost estimate for rule implementation concluded the EPA cost estimates for rule
implementation are more than an order of magnitude low and that many proposed emission
estimation methodologies are cost-prohibitive. EPA has not defined or provided inventory or
data quality objectives to justify these extensive costs.
Response: Please see EPAs response to EPA-HQ-OAR-2009-0923-1167-11 for a response to
the commenters cost estimate. In addition, EPA disagrees with the commenters statement that
the Agency did not discuss data quality objectives. See the Economic Impact Analysis, in
particular Section 3.4, for relevant discussion.
1037

While Devon strongly supports environmental and public health protection, we are highly
concerned that this proposed reporting rule will impose an unnecessarily high cost to our
industry for minimal air quality benefit. As a matter of fact, our internal economic impact
analysis of this proposed rule indicates that the estimated cost to Devon, without contingency, is
$28 million. The cost to Devon alone exceeds what the EPA has estimated ($27.7 million) for the
entire US Oil and Gas industry!
Imposing costs of this magnitude on an operator results in an equivalent reduction of capital
investment which results in fewer wells drilled, less clean burning natural gas produced, fewer
local jobs, and less revenue returned to the US Treasury from severance taxes and mineral
royalty payments.
Response: EPA is unable to evaluate the commenters compliance cost estimate of $28 million
because insufficient documentation was provided to explain how this figure was calculated. For
example, the commenter did not provide assumptions for the labor rate(s), labor hours, or a
detailed labor description used to calculated the estimated burden. For information about the
assumptions and costs estimates for todays final rule, see the Economic Impact Analysis (EPA
HQ-OAR-2009-0923). For EPAs responses to other comments about the onshore production
cost estimates, see comments EPA-HQ-OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009
0923-1151-105. For information about the sampling and analysis of hydrocarbon liquids in
tanks, see EPA-HQ-OAR-2009-0923-1011-21; see EPA-HQ-OAR-2009-0923-1151-129 for
information about reporting of produced water emissions.

The proposed rule will be unreasonably burdensome and costly to small businesses, and could
cause many marginal wells to become uneconomic and abandoned.
Response: Please see Section 5 of the Economic Impact Analysis (EPA-HQ-OAR-2009-0923)
and EPAs response to EPA-HQ-OAR-2009-0923-0053-1 for a response to this comment.

Organization: BP America, Inc.
Commenter: Karen St. John

Burden and Timing for Onshore Petroleum and Gas Production
1038
The expansion of the MRR to include the entire natural gas supply chain, from dispersed wells
and small facilities to distribution meter stations and city gates; tremendously expands the scope
of the rule in terms of number of sites, source types, and monitoring/calculation methodologies.
BP believes that the cost and burden of the rule depicted in the Economic Impact Analysis (EIA)
for the newly included Onshore Petroleum and Gas Production (OPGP) category is significantly
understated and should be updated prior to finalizing the rule with the OPGP category included.
Our lower 48 Onshore Petroleum and Gas Production operations, composed of some 13,868
wells along with several hundred ancillary small compressors and facilities, have projected the
initial 2010-2011 burden of the rule at some $23.6 MM with annual reoccurring costs of some
$17 MM excluding the costs for quarterly sampling and analysis of hydrocarbon liquid and
produced water for dissolved CO2. Inclusion of the costs estimated for these two source types
boosts this total to more than $40 MM. This suggests that the burden estimate of $27.7 MM in
the EIA is very low. This further suggests that the amount of work necessary to implement the
rule requirements also expands by about the same ratio and cannot be accomplished in the time
frame contemplated in the proposed rule.
To address the balance of burden and cost with emission coverage and the program goals of
informing policy along with adequate time for implementation, BP suggests the following:
- EPA should conduct a re-evaluation of the Economic Impact Analysis and rationalization of the
rule requirements, rule approaches, burden, and costs vs. the amount of emissions covered for
additional public comment before finalizing the GHG reporting rule for the Onshore Petroleum
and Natural Gas Production sector.
Response: EPA disagrees with the comment. EPA is unable to evaluate the commenters
compliance cost estimate of $40 million because insufficient documentation was provided to
explain how this figure was calculated. For example, the commenter did not provide
assumptions for the labor rate(s), labor hours, or a detailed labor description used to calculated
the estimated burden. Regarding the American Petroleum Institute (API) cost estimates
referenced by the commenter, EPA has provided detailed responses to API explaining why their
estimates are too high; see EPA-HQ-OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009
0923-1151-107 for further details. EPA-HQ-OAR-2009-0923-1151-107 identifies source
categories discussed in each comment. In addition, see EPA-HQ-OAR-2009-0923-1151-5 for
EPAs response to comments that it re-evaluated the Economic Impact Analysis and basis for
rule scope and requirements. EPA adjusted the analysis in response to comments. For example,
EPA included determination costs for non-reporters that will have to expend resources to
determine whether or not to report. In addition, EPA updated the burden to reflect changes in
todays final rule, such as monitoring of emissions sources that now have equipment thresholds
for monitoring and labor hours for monitoring of equipment leaks in onshore production.
Finally, regarding concerns about the time to implement the rule, see Section II of the preamble.
1039

Economic Impact Analysis and Burden
BP participated in APIs cost analysis of the proposed rule and fully supports APIs analysis and
conclusions - particularly for the newly included Onshore Petroleum and Gas Production
(OPGP) category. Our analysis of the costs for the wells and facilities we operate suggest that
APIs analysis is correct and that EPAs projection of first year costs of $27.7 MM (preamble
Table W-5) for this category is too low by more than an order of magnitude.
Response: Please see EPAs responses to comments EPA-HQ-OAR-2009-0923-1151-89
through EPA-HQ-OAR-2009-0923-1151-105 for a response to this comment.

EPA Underestimates the Cost of Complying with Subpart W.
EPA's cost estimate for Subpart W compliance is many orders of magnitude too low considering
the hundreds of thousands of sources for which our industry would need to document and report
emissions. EPA estimates that the costs to be incurred by onshore producers for the first year of
compliance will be approximately $24,000 per reporting unit. See Table W-10, 75 Fed. Reg. at
18,629 .
To confirm the accuracy of EPA's estimate, Chesapeake conducted its own analysis for an
individual reporting basin. 112 Chesapeake's analysis assumed an individual reporting basin where
all operations are subject to reporting. Chesapeake reviewed the calculations and reporting
requirements in the proposed Subpart W to determine the emissions from the following sources:
- Single-well pads
- Multi-well pads
- Glycol Dehydrators
- Reciprocating Compressors
- Completion Operations
- Workover Operations
Chesapeake's estimate was developed using estimates of labor rates, level of effort, and
equipment rental costs. Based upon our analysis, Chesapeake calculated a first-year compliance
cost of nearly $10.5 million for one representative basin - an estimate well over 400 times higher
than that determined by EPA.
112
Chesapeake acknowledges that the actual compliance costs at a given basin will vary greatly depending on the
equipment operated and levels of activity.
1040
Response: EPA has reviewed the commenters costs, including the information they submitted
to the docket and identified as confidential business information. EPA is unable to evaluate the
commenters estimate because insufficient documentation was provided to explain how the
figures were calculated. The commenter has developed source categories such as single-well
pads and completion operations and assigned reporting costs to each source category. However,
the commenter did not disclose which emission sources to which these costs apply. In addition,
the commenter did not disclose details about the labor rate, level of effort, and equipment rental
costs used to estimate the compliance costs for each source category. The remainder of this
response presents EPAs consideration of the information provided by the commenter.
Single and Multi-Well Pads
EPA cannot respond to the commenters costs for single-well and multi-well pads because
insufficient details were provided. Although, the commenter has disclosed the equipment at each
single and multi-well pad, they did not specify EPA is uncertain what data points are being
collected under these categories. To accurately respond to the commenters compliance cost for
single and multi-well pads, EPA would need to know, what data is being collected and how the
data is being collected. See response to EPA-HQ-OAR-2009-0923-1151-5 for discussion about
changes EPA made to todays final rule to reduce burden for onshore production, such as
requiring operators to count their major equipment instead of each individual component.
Glycol Dehydrators
Please see EPA-HQ-OAR-2009-0923-1151-89 90 for EPAs response to comments about the
costs for glycol dehydrator compliance cost estimate and details about EPAs estimates.
Reciprocating Compressors
Please see comment EPA-HQ-OAR-2009-0923-1063-1 for a response to comments about the
costs for reciprocating compressors.
Completion and Workover Operations
The comment did not distinguish between unconventional and conventional completions and
workovers. Therefore, EPA cannot evaluate the commenters compliance cost estimate for
completion and workover operations. Please see comment EPA-HQ-OAR-2009-0923-1151-97
for clarification about how reporting emissions from well completions and workovers with and
without hydraulic fracturing. Also, please see Section 4 of the Economic Impact Analysis (EPA
HQ-OAR-2009-0923) for EPAs compliance cost estimates for conventional and unconventional
well completions and workovers.
Overhead, Quality Control, and Reporting Costs
Although EPA agrees with the commenters overhead, quality control, and reporting costs per
basin, EPA disagrees with the commenters company-wide IT costs. Please see comment EPA
HQ-OAR-2009-0923-1151-106 for discussion about EPAs consideration of the costs for
company-wide data management systems.

1041

First, the proposed requirements will be more costly to implement than EPA has estimated.
Because they apply to vastly more industry activities and sources, EPAs cost estimates for
impact on the industry are too low by an order of magnitude
Response: Please see EPAs responses to comments EPA-HQ-OAR-2009-0923-1151-89
through EPA-HQ-OAR-2009-0923-1151-105 for a response to this comment.
Organization: Gas Processors Association
Commenter:

EPAs estimated cost impact appears to be at least an order of magnitude low.
GPA is evaluating the economic impact of the proposed Subpart W and will compare the results
to the analysis provided by EPA. However, preliminary indications are that the EPAs cost
estimates are at least one order of magnitude low. EPA has failed to recognize the full impact of
requiring reporting at all facilities and pipelines in the natural gas gathering and processing
sector.
To demonstrate the disparity in EPAs cost estimate, the following cost estimates from one GPA
member company are presented. These costs are only for this one company, an independent
gathering and processing company with no production. Their estimated first year costs for initial
implementation of the proposed Subpart W are approximately $80 million and their ongoing
annual costs for recording, measuring, calculating, and reporting are estimated to be $9.5 million.
To emphasize, these are the estimated costs for only one GPA member company to comply with
the proposed Subpart W. This compares to EPAs estimate for all companies in all segments of
the natural gas industry covered by Subpart W of $56 million for the first year costs and $21.4
million for ongoing annual costs.
Response: EPA disagrees that the proposed rules cost are at least an order of magnitude low.
EPA is unable to evaluate the commenters estimate for one facility$80 million year one and
$9.5 million in subsequent yearsbecause insufficient documentation was provided to explain
how these figures were calculated. For example, the commenter did not provide its assumptions
for the cost of each monitoring activity per source or how these were extrapolated.
However, EPA has made several clarifications and revisions that will reduce the burden on
processing reporters, as described in Preamble Section II.F. For instance, EPA has decided not
to include gathering lines and boosting stations as an emissions source in subpart W at this time,
see comment response EPA-HQ-OAR-2009-0923-1206-13. Second, todays final rule clarified
that reporters may conduct an annual measurement of each compressor in the mode as it exists at
1042
the time the annual measurement is taken, with shutdown depressurized mode monitored for
each compressor at least once every three years. Please see Compressor Modes and Threshold
in EPA-HQ-OAR-2009-0923. Third, todays final rule does not require tank emission
calculations for natural gas processing.
See also the Economic Impact Analysis (EPA-HQ-OAR-2009-0923) for more information about
EPAs estimates of todays final rules cost for the average natural gas processing facility.
Finally, this commenter submitted additional comments and cost estimates to the docket; see
EPA-HQ-OAR-2009-0923-1206-25 for EPAs response.
Commenter: Pamela Lacey
For purposes of these comments, we will assume that optical gas scanning refers to infrared
cameras, which cost between $80,000 and $100,000 each to purchase, or about $2,500 per week
to rent. We also will assume that EPA means to include industrial and commercial customer
meters and regulators, but not residential meters, within the term above ground M&R station.
To evaluate the burden of Subpart W, AGA conducted a survey of our members to determine
how many above ground M&R stations would be affected by the proposal. The results are
somewhat difficult to interpret, because it became apparent that our members are very confused
by EPAs undefined term. Some thought it included residential meters, other thought it only
included industrial and commercial meters, and all questioned whether metering and regulating
for smaller commercial customers such as fast food restaurants and medium box stores would be
included. The survey will need to be refined, once EPA provides clarification regarding the
scope of this term, but the current survey indicates that our member LDCs have an average of
19,000 M&R stations per company that would be affected by this proposed annual infrared
camera leak survey requirement. Our members estimate that they can visit less than 10 M&R
stations per day to perform a leak survey. As a result, the Subpart W leak survey proposal could
easily require each company to purchase or rent 50-100 infrared cameras to get the survey work
done within about a two month timeframe each year, beginning in 2011. As a result, it could
easily cost a single LDC $4,000,000 to comply with this leak survey requirement in 2011. In
addition, there are not enough cameras or trained operators to supply this demand, so market
forces would likely increase costs further. EPA asserts that LDCs would be able to outsource
their optical scanning needs by contracting with consulting firms. However, there are also not
enough cameras or trained operators within consulting firms to serve this level of demand.
EPA estimates in the preamble to the proposed rule that the entire natural gas distribution sector
all affected LDCs in the United States would incur only $1.6 million as the first year cost for
reporting fugitive and vented emissions under all provisions of the Subpart W proposal. In
contrast, we estimate that a single average AGA member LDC could incur about $4 million in
the first year to comply with a single provision in the Subpart W proposal. EPA has clearly
misunderstood the impacts of this rule and has vastly underestimated the cost of providing the
information EPA seeks.
1043
Response: As described in Section III.B of the preamble to todays final rule, EPA has reviewed
a number of comments concerning the proposed rules cost of compliance and disagrees these
costs are vastly underestimated. Regarding EPAs response to the commenters assumption that
the rule would apply to customer meters, the requirements do not apply to customer meters
(industrial, commercial, and residential meters) or to farm taps. See preamble Section II.F for a
complete discussion about the meters EPA intended to include in the rule. EPAs response to
EPA intended to include in the rule.
EPA finds the commenters estimates of the costs for each camera to be reasonable. EPA
estimated the overnight capital cost for each camera to be $100,000, which is at the upper bound
of the GPA cost range of $80,000 to $100,000. However, EPA disagrees with the commenters
assumptions about the extent of requirements for leak detection. EPA estimates that each LDC
will be required to monitor between 32 and 215 metering and regulating stations, a small fraction
of the 19,000 metering and regulating stations assumed by this comment. As discussed in
Preamble Section III.B and in comment EPA-HQ-OAR-2009-0923-1016-11, the commenter
based its cost estimate on a much higher number of metering and regulating stations because it
assumed that Subpart W leak detection would apply to all customer meters as well as meter and
regulator stations at which custody transfer does not occur.
Regarding the commenters concerns about the optical scanning needs under the rule, EPA has
revised todays final rule and is including the option to use Method 21 and infrared laser beam
illuminated instruments to detect leaks for sources that are accessible, as described in Preamble
Section II.F.
1044

Organization: Cardinal Engineering, Inc.
Commenter: Kristine D. Baranski

Although the commenter appreciates that there are concerns with obtaining comprehensive data,
the direct inspection cost involved with optical gas imaging is extremely high .
The cost is high when one considers that the output is qualitative (as compared with a ppm
"leaking" threshold obtained via the OVA/TVA monitoring methodology as currently practiced
in LDAR programs) .
Response: EPA agrees with the comment that in certain situations, using handheld leak
detection devices may be more cost effective. Hence, EPA has revised todays final rule and is
including the option to use Method 21. Please see Section II.E of the preamble for further
details.

Organization: Canadian Gas Association
Commenter: Michael Cleland
In the rule preamble (tables W-S and W-6, p. 89), EPA estimates costs first year costs for LDCs
to implement the rule at $0.07 per tonne in the first year, dropping to $0.04 per tonne in
subsequent years. Based on Canadian LDC experience in monitoring and reporting, we are
unable to come anywhere near to these very low amounts.
Using Canadian data we have come up with some estimates that suggest much higher costs will
be incurred.
For example, based on field data and analysis conducted since 1995, the default emission factor
for a gate station equates to 22.4 t CO2e per station. 113 At $0.07/tCO2e, the per gate station
measurement budget equates to $1.57.
Through the CGA's CEPEI group, the downstream natural gas industry has collaborated on
compiling a national GHG inventory for some years. Based on the most recent year that national
data were compiled (the 2007 inventory year), fugitive emissions from the distribution sector in
Canada totaled 890,000 t CO2e, which represents 0.1 % of Canada's national inventory in 2007.
113
Methodology Manual: Estimation of Air Emissions from the Canadian Natural Gas Transmission, Storage and
Distribution System prepared for Canadian Energy Partnership for Environmental Innovation by Clearstone
Engineering, calculation Form 3.2.1+10, p. 49 (September, 2007); the emission factor has been converted to tones
CO2e based on gas composition data from Ontario.
1045
This amount represents the emissions reported by Canada's 12 main LDC operations 114 , an
average of 74,200 t CO2e per company. At $0.07 pert CO2e, this equates to an average cost per
company of $5,200 , an amount which could not reasonably cover the field survey work,
"administrative costs reviewing the reporting rules, training personnel, documenting emissions
data and emissions estimates, approving the submission to the EPA, submitting reports and
maintaining records 115 that EPA has included in its cost estimates.
Response: Overall, EPA disagrees that the Agency underestimated the rules compliance costs,
and the distribution segment in particular, as described in the response to comment EPA-HQ
OAR-2009-0923-1016-4. The remainder of this response discusses details unique to this
comment.
First, EPA disagrees that each reporters per gate station measurement budget equals $1.57.
While EPA estimated an average compliance cost of $0.07 per metric ton CO2e for the
distribution segment in the proposed rule, this average does not represent the costs associated
with individual activities within that segment. For instance, the per metric ton CO2e cost will be
higher when leak detection is conducted; the per metric ton CO2e cost will be lower for
monitoring activities that rely on a population emission factor is required.
Furthermore, EPA disagrees with the commenters estimated average cost per company of
$5,200. EPAs cost estimates are based on the analysis of the distribution segment in the United
States; application of EPAs estimates to jurisdictions outside the United States may not be
appropriate or accurate. For instance, in the proposed rule, EPA estimated the average emissions
per LDC to be approximately 159,000 metric tons, which is more than twice as high the
commenters estimated average emissions per LDC. EPA analyzed the costs of compliance for
natural gas distribution and has revised the estimated cost of compliance in todays final rule for
the average reporter to be $0.09 per metric ton and $13,854 per LDC for the first year. Further
comment on this topic is available in Section III.B, of todays preamble.
NGSA believes that the proposed Subpart W will cause formidable financial and administrative
burdens without contributing appreciably to the coverage or accuracy of emissions monitoring
under the Mandatory Reporting Rule. In these comments, we discuss several ways in which the
proposed Subpart W should be modified to better reflect the realities of monitoring GHG
emissions in the natural gas production sector.
Response: EPA does not agree with this comment; please see EPA-HQ-OAR-2009-0923-1004
4 for a response to this comment.
114
These companies represent almost the entire LDC sector operations in Canada.
115
EPA Subpart W preamble - p. 8788.

1046

The proposed Subpart W still includes highly labor-intensive and equipment-intensive
monitoring methods for our sector requiring population counts of thousands of individual
components (many of which are trivial sources of emissions) at hundreds of thousands of
wellheads and thousands of miles of gathering pipelines, as well as direct measurement of
equipment such as rod packing vents, unconventional well completions and workovers, and wet
seal degassing vents. Certain units, such as acid gas removal (AGR) stacks, are also not equipped
with the metering or monitoring equipment needed to fully comply with the proposed Subpart
W; numerous units will, in many cases, need to be taken offline to install the required equipment.
Considering that our sector contains approximately 475,000 of wellheads alone, these measures
will require massive investments in data collection on facilities (and components within
facilities) that are often negligible sources of GHG emissions.
Under any reasonable set of assumptions, the cost of complying with the above requirements will
far exceed EPAs estimate of only 27.7 million dollars for the entire onshore petroleum and
natural gas production sector. Indeed, the American Petroleum Institutes (API) review of the
rule indicated that the costs of the proposed Subpart W to the onshore petroleum and natural gas
production sector alone could reach 1.7 billion dollars, or 8.44 dollars per ton CO2-e of GHG
emissions. If correct, this estimate would cause the cost of the proposed Subpart W to approach
the projected near-term cost of a mandatory cap-and trade program even though the sole
purpose of the Mandatory Reporting Rule is to collect policy-relevant information on GHG
emissions, not impose emission controls. In addition, this figure would far exceed EPAs
estimated cost-per-ton for any other sector included in the Mandatory Reporting Rule.
NGSA believes that EPAs proposed basin-level approach fails to achieve an appropriate balance
between the need for GHG data with the practical and economic burdens associated with the
reporting burden imposed by the proposed Subpart W.
Response: EPA does not agree with the comment. Regarding the American Petroleum
Institutes estimate of the costs, as cited by the commenter, please see EPAs response to EPA
HQ-OAR-2009-0923-1151-107, which provides a summary of the cost breakdown. Regarding
the commenters statements about requirements for population counts, please see EPA-HQ
OAR-2009-0923-1151-89.
For a response to the comment on direct measurement of equipment such as rod packing vents,
well completions and workovers with hydraulic fracturing, and wet seal degassing vents, see
EPA-HQ-OAR-2009-0923-1151-94 and EPA-HQ-OAR-2009-0923-1151-97.
For a response to the comment that certain equipment, such as acid gas removal vents, are not
already equipped with meters or monitoring devices, see EPA-HQ-OAR-2009-0923-1024-26.
Regarding the commenters statement about the basin-level approach, see EPA-HQ-OAR-2009
0923-1151-5.
1047

Further, the agency is proposing to require annual leak surveys using, costly and unnecessary
optical scanning equipment that does not result in improved leak detection beyond the leak
detection currently conducted by gas utilities under existing federal and state pipeline safety
regulations. This proposal would impose billions of dollars in cost on gas utilities and their
customers rivaling costs under a cap and trade program without reducing emissions. All this
effort would provide no better picture of GHG emissions from this segment than is currently
available in the annual EPA GHG Inventory, because LDCs would have to use outdated
emission factors that tend to seriously overstate GHG emissions from natural gas distribution.
Response: As described in the response to comment EPA-HQ-OAR-2009-0923-1152-4, EPA
disagrees that the costs associated with the rule would be on the order of billions of dollars.
Regarding the commenters concerns about the optical scanning needs under the rule, EPA has
revised todays final rule to allow alternative options. See preamble Section II.F for details
about the alternative leak detection options.
EPA also disagrees that the information gathered is duplicative with the EPA GHG Inventory.
As described in EPA-HQ-OAR-2009-0923-1016-11, todays final rule will provide an estimate
of actual emissions, as opposed to "potential of emissions. In addition, the proposed rules
leaker emission factors for natural gas distribution are based on recent emission measurement
studies conducted by Clearstone Engineering, and not the U.S. GHG Inventory. Further
discussion on this topic is provided in Greenhouse Gas Emissions from the Petroleum and
Natural Gas Industry: Background TSD (EPA-HQ-OAR-2009-0923-0027).

Total Cost for Process Emissions
In consideration of the above discussions, it is Southwests opinion that the Process Emissions
cost estimate described in the preamble of the proposed rule is significantly underestimated. The
published cost estimate for the natural gas distribution sector is estimated to be $1,680,000 for
start up and $600,000 for annual costs thereafter. Southwest respectfully suggests that these costs
are more likely to be $1,680,000,000 and $600,000,000, respectively.
Response: Note: See comment EPA-HQ-OAR-2009-0923-1020-5 for the text that precedes this
excerpt from the commenters letter.
1048
EPA has reviewed the proposed rules cost estimates and disagrees that the Agency significantly
underestimated the costs for the natural gas distribution segment. Based on the limited
information that the commenter provided regarding the basis for their estimate, EPA has
determined that the commenter assumed that leak detection is required for all regulator stations.
Please see the response to comment EPA-HQ-OAR-2009-0923-1016-11 for detailed discussion
about why this assumption is inconsistent with EPAs intent.
EPA is unable to further evaluate the commenters estimates ($1.68 billion and $600 million)
because insufficient documentation was provided to explain how these figures were calculated.
For example, although the commenter specified their assumption about the cost of infrared laser
detector instruments, they did not identify the other assumptions and calculations underlying its
estimate, such as labor involved. EPA estimated the capital cost of an optical gas imaging
instrument to be $100,000, similar to the cost estimated by the commenter. However, the
commenter assumed they would purchase the instrument and did not specify whether they
considered the lowest cost option that EPA analyzed. EPA assumed that reporters would use
contractors to conduct emissions detection activities; the contractors would purchase the
instrument and pass on the costs to as many facilities as they can provide service to each year
and it was unclear if the commenter agreed with these assumptions or not because they did not
indicate they considered this option. EPA modified this analysis in the final Economic Impact
Analysis based on the revision to todays final rule that allows alternative leak detection
methods. For details about the alternative leak detection methods, see preamble II.E; the
response to comment EPA-HQ-OAR-2009-0923-1016-11 provides details about the associated
costs.
Finally, the commenter provided an estimate to conduct a component inventory that appears to
have contributed to the total cost estimate. See EPA-HQ-OAR-2009-0923-1016-11 for a
response to assumptions about component inventories. .

Discussion of Implementation Costs
Based on information provided in EPAs Economic Impact Analysis (EIA), the annual cost per
affected facility for measurement and monitoring in subsequent years (i.e., after the first year of
reporting) is approximately $16,000. This total includes stationary combustion source reporting
that will be required upon Subpart W promulgation. For small to average-size facilities without
complicated measurements, INGAA generally agrees with the estimated subsequent year costs.
However, the EPA cost analysis did not adequately account for all required and potential costs,
including the following:
Additional first-year reporting costs. First-year per-facility costs presented in the EIA are only
about $1,700 more than subsequent year per-facility costs. Additional first-year tasks would be
required, including a more detailed assessment of applicability and requirements; monitoring
1049
plan development, including standard operating procedures for measurements; equipment

calibrations; measurement criteria definitions; identification of vent access measurement issues
and sample port installations; calculation and recordkeeping documentation development (e.g.,
spreadsheets and database support); operator and technician training; data review/QA-QC
procedures; reporting tool development and integration with EPAs pending web-based reporting
system; and, additional management oversight. INGAA understands that some tasks would be
company-wide and shared by numerous facilities. However, implementing a program to comply
with a rule of this breadth and complexity would likely entail costs an order of magnitude or
more greater than EPAs estimate for the initial reporting year.
Three-mode testing costs for reciprocating compressors. As discussed in Comment V, emission
measurements must be conducted for each compressor in each of three operating modes that
occurs within the year, including operating, standby-pressurized, and not operatingdepressurized. This requirement will be considerably more complex than simply testing each
compressor in a single as-found condition. One approach would be to cycle each compressor
through the three modes during a test team visit. However, this would complicate compressor
station operation (i.e., coordination with dispatch), possibly disrupt gas transmission service, and
would most likely create an unnecessary venting event for each compressor that requires a
blowdown which wastes product and increases GHG emissions. Alternatively, a test team could
be placed on standby while each compressor naturally cycles through the three modes or could
revisit the compressor station numerous times; or, operators could be trained to perform
measurements when the opportunity arises. Each of these approaches creates its own
complications and expense. It does not appear that the EPA analysis accounts for the costs
associated with the complexities of three-mode testing.
Testing complexity for inaccessible vents. Many compressor vents, such as reciprocating rod
packing and centrifugal wet seal oil degassing, are located above rooflines since facilities were
not constructed with vent access in mind. Such vents would be deemed inaccessible in many
cases under typical LDAR programs. Accessing these vents to measure flowrates typically
requires a manlift and could require assessment and implementation of additional safety
precautions (e.g., by-passing high-pressure gas lines, avoiding personnel exposure to an
emergency blowdown event, etc.). These requirements will add time and expense to
measurements, especially for three-mode testing of reciprocating compressors.
Small facility testing to validate non-reporting status. If a method to screen facilities for rule
applicability (i.e. 25,000 metric ton CO2e threshold) is not defined or if the screening method is
too conservative, then the number of facilities required to conduct first year measurements will
increase. For the 25,000 metric ton threshold, EPA estimates that 1,145 of 1,944 transmission
compressor stations and 133 of 397 natural gas storage facilities will be required to report GHG
emissions. Without an accurate applicability screening method, first- year testing costs could
increase significantly and may encompass all facilities to confirm emissions from facilities
required to report and validate status of smaller facilities not subject to the rule
Response: EPA determined that screening tools would facilitate reporting determinations and
plans to make such tools available. Accordingly, EPA has updated the Economic Impact
1050
Analysis to better account for reporting determinations and expected use of screening tools; see
EPAs complete response in Section III.B.2 of the preamble to todays final rule.
On the other hand, EPA disagrees that the first year cost estimates should be an order of
magnitude greater. As discussed in the Economic Impact Analysis and Preamble Section III, the
inclusion of reporting costs increased EPAs estimate of total private sector costs compared to
the April 2010 analysis, but the order of magnitude remained the same. EPA included similar
first year costs to the commenter in the proposed rule, such as registration; regulation review,
monitoring plan development; equipment; monitoring and measurement, training,
documentation, reporting, auditing, and archiving. As noted above, determination costs have
been added to this list I todays final rule, determination costs have been added to this list.
EPA did not intend for reciprocating or centrifugal compressors to be taken offline in order for
reporters to collect the operational mode data required under subpart W. Therefore, as described
in Preamble Section II.F, EPA has clarified the todays final rule to allow reporters to conduct an
annual measurement of each compressor in the mode as it exists at the time the annual
measurement is taken, with shutdown depressurized mode monitored for each compressor at
least once every three years. For more detail, please see Compressor Modes and Threshold in
the docket (EPA-HQ-OAR-2009-0923).
EPA considered the comment concerning monitoring of inaccessible compressor vents, and
disagrees with the comment on testing complexity because the proposed rule allows installation
of a port on the vent line for insertion of a temporary meter, and these ports can be installed at
ground level. Additionally, in todays final rule, reporters may now use acoustic leak detection
devices to estimate through-valve leakage from unit valves and blowdown valves. Please see
Section II.F of the preamble.
Applicability Determination Costs and Burden
In discussing the Proposed Rule implementation schedule, EPA assumes that many reporting
entities already have GHG monitoring capability due to the requirements of other air quality
programs. This assumption is not valid for natural gas transmission and storage systems, which
have never been subject to direct measurement of vented emissions or fugitive emissions
monitoring, as required in the Proposed Rule. Natural gas transmission and storage facilities do
not have currently-installed mechanisms or data systems for monitoring and measuring fugitive
or most vented fugitive emission sources as called for in the Proposed Rule. For INGAA
members, the task of determining whether GHG reporting is required for specific facilities under
Subpart W would represent a significant departure from current practices, requiring considerable
time and resource allocations.
In reviewing EPA cost estimates, INGAA generally agrees with the per facility monitoring costs
for years subsequent to the initial year. However, that cost does not consider important logistical
factors such as vent access safety issues, three-mode testing, and availability of service
1051
providers, which will escalate per-facility costs. In addition, EPA cost estimates do not consider
costs associated with measurement and monitoring at smaller facilities to determine their need to
report; therefore, cumulative costs are significantly underestimated. For natural gas transmission
and storage, EPA estimates that 59% of 1944 compressor stations and 34% of 397 underground
storage facilities will exceed the reporting threshold [see 75 FR 18618], and approximately 1060
facilities (or 45%) would not report for these two industry segments. However, as noted above,
absent a screening method, these facilities would still be faced with vent measurement, leak
monitoring, and population counts for pneumatic devices and storage wellhead components to
estimate facility GHG emissions and document that total facility emissions are less than the
threshold, in essence to provide a negative determination relative to the 25,000 metric ton
CO2e threshold. Thus, with an additional 45% of facilities that do not report still required to
conduct monitoring (e.g., leak surveys) and direct vent measurement, the cumulative costs would
be nearly double EPAs estimate.
In addition, the number of available service providers and qualified technicians is limited, and
monitoring and measurement are required for many segments in Subpart W. Thus, the added
need to monitor at smaller facilities would exacerbate this shortage, and market pressure could
escalate per-facility costs. This shortage could also compromise the ability of operators to meet
the Proposed Rule schedule and is one of several factors contributing to the need for a phased
approach for reporting. In addition, the Proposed Rule compromises compliance certainty due to
the uncertainty associated with GHG estimates for fugitive emissions and some vented sources.
The Proposed Rule essentially requires that Subpart W emission estimation methods (monitoring
and direct measurement) be applied to every natural gas transmission compression facility and
natural gas storage facility, every year. These requirements would not only increase the cost of
Subpart W, but also negate the administrative and cost advantages that EPA sought to achieve by
selecting a 25,000 metric ton CO2e threshold. In conclusion, a screening method that provides
reasonable compliance certainty is needed to avoid unnecessary compliance risk, implementation
complexity, and financial burden
complete response in Section III.B.2 of the preamble to todays final rule. In addition, todays
final rule allows reporters to request the use of best available monitoring methods under certain
conditions, such as lack of service providers and qualified technicians; please see the response to
EPA-HQ-OAR-2009-0923-1011-27 for further details. Also see response to EPA-HQ-OAR
2009-0923-1039-25 that addresses the commenters concerns about vent access safety issues and
three-mode testing.
Impact on Local Distribution Companies
1052
Several commenters have indicated that the proposed Subpart W will place an onerous burden on
local distribution companies (LDCs), with a financial impact as great as $4,000,000 per LDC.
While we agree that the coverage of LDC systems under Subpart W is in need of clarification,
we do not believe that EPA intended the coverage of LDC systems to be as sweeping as these
commenters suggest, and believe that EPA can proceed to require OGI for emission detection at
city gate stations and above ground district regulators. In addition, we believe it is advisable for
EPA to require OGI emission detection at certain underground pipeline main facilities and large
customer metering and regulating stations, albeit on a phased in timetable.
1. Proposed LDC Coverage of Subpart W is Manageable and Cost-Effective. Although the
proposed definition of LDC facilities, which refers broadly to above ground meter regulators
and gate stations, 116 is ambiguous, we do not believe EPA intended for the proposed Subpart W
to have such broad coverage as to impose an OGI inspection requirement on residential meters
and small commercial establishments. As EPA explains in the preamble to the proposed rule:
Distribution system CH4 and CO2 emissions result mainly from fugitive emissions from
above ground gate stations (metering and regulating stations), below grade vaults
(regulator stations), and fugitive emissions from buried pipelines. 117
EPAs view that gate stations, below grade regulator stations, and buried pipelines are the main
contributors to GHG emissions from LDC systems is supported by the data provided in
Appendix A of the Technical Support Document accompanying the proposed rule. The TSD,
and the preamble, do not mention requiring OGI detection or emission factor estimates for any
customer-specific metering facilities.
Assuming that EPA intended for Subpart W to have this more limited scope, we believe that the
proposed rule could be easily implemented at reasonable cost. Based on our inquiries with a
local distribution company serving a large city in the Northeast, the number of above-ground city
gates and district regulator stations should be manageable even for a large metropolitan area.
The utility we consulted has 20 city gate terminals which meter and reduce the pressure from
transmission pipeline(s). The utility also has 150 district regulators which operate downstream
from the city gate terminals and reduce pressure and re-distribute the gas to the lower pressure
gas distribution system. These stations are housed in underground spaces, underground manhole
vaults and in small above ground buildings and sometimes within the gate station itself. The
system in our survey has only 6 district regulators below ground.
A summary of estimated costs to survey these locations is presented below:
116
Proposed 40 C.F.R. 98.230(a)(8).
117
75 Fed. Reg. at 18,617

1053
Based on this survey, we suggest that the costs associated with using OGI at gate stations and
district regulators is nowhere near the multimillion dollar annual expense suggested by some
commenters. Moreover, finding and repairing leaks would yield savings for LDCs and ratepayers
possibly enough savings to more than pay for the consulting services or equipment and
program implementation costs.
FLIR Systems also disagrees with EPAs assumption that emissions from underground vaults
cannot be detected using OGI, and must therefore be estimated using population emission
factors. FLIR Systems has successfully deployed OGI at underground vaults by lowering OGI
equipment through the same access points used for maintenance and repair of the systems within
these vaults. Given that below-grade regulator stations are both manageable in number and
capable of being monitored using OGI, FLIR Systems recommends that EPA consider proposing
the use of OGI for detection of fugitive emissions from underground vaults.
Response: EPA agrees that the costs for optical gas imaging are well below the multimillion
dollar estimates. EPA also generally agrees with the number of station locations per LDC;
however, the Agency did not intend to require surveys for non-custody transfer stations and has
clarified todays final rule (see preamble II.F). In addition, the comment did not provide
sufficient information for EPA to respond in detail to the number of hours required per station.
For instance, while the commenter estimates four hours to survey an above ground gate station,
including 45 minutes travel time between stations and time to prepare reports, it is unclear how
much time is allocated to survey each station compared to preparing reports. In addition, the
data presented by the commenter suggest that the cost for consultants is $375 per hour. EPA
disagrees with this hourly rate; in todays final rule, EPA has applied labor rates that are
significantly lower than that assumed by the comment: from $55.20 per hour to $101.31 per
hour, as described in Section 4 of the Economic Impact Analysis (EPA-HQ-OAR-2009-0923).
In todays final rule, EPA has revised the estimated cost of compliance for the average reporter
to be $0.09 per metric ton and $13,854 per LDC for the first year. See EPA-HQ-OAR-2009
0923-1036-11 for a detailed discussion about EPAs assumptions and cost estimates for leak
detection surveys.
Finally, EPA disagrees with the commenters characterization of the Agencys assumptions in
the proposed rule for leak detection in metering and regulating underground vaults. EPA did not
assume that equipment leak emissions from underground vaults cannot be detected using optical
1054
gas imaging instruments. Rather, EPA determined that underground vaults can be difficult to
access, and equipment leak emissions detection at underground vaults can be burdensome.
Therefore, todays final rule requires the use of population emission factors.
Organization: Southern Union Company
Commenter: Charles Wait

SU wishes to emphasize a few of the major issues that our company faces. First, SU urges EPA
to consider the INGAA proposed screening method. SU believes that only 60-65 percent of its
stations have greenhouse gas (GHG) emissions greater than the reporting threshold. However,
absent a screening methodology in the final rule, SU feels obligated to devote precious resources
to monitoring stations to prove a negative applicability. These include stations where total GHG
emissions could be as low as 5,000 mT CO2e.
complete response in Section III.B.2 of the preamble to todays final rule.
Organization: New Mexico Gas Company
Commenter: Curtis J. Winner
Component count of fugitive emissions sources - Section 98.236.c.19 is confusing. Does this
section require us to do a component count of all our facilities regardless of whether they are
leaking? If this is EPA's intention, this would be very burdensome and costly. Each of our
distribution meters for example has approximately 100 connectors alone. This data is not
currently compiled in any database. Therefore, to comply with this reporting requirement we will
have to conduct a survey of all our meters (270 to 500,000 depending on the definition) and
count all the components. We would need to employ hundreds of people for several months to
accomplish this. Why is this data required? And what exactly does EPA want us to report?
Response: Todays final rule does not require a component count regardless of whether they are
leaking; rather, todays final rule requires the counting of only leaking components. Also, the
rule does not require monitoring of all meters from natural gas distribution reporters; only
leaking components at above ground city gate stations at custody transfer have to be counted and
reported to EPA. See preamble Section II.F for a complete discussion about the meters EPA
intended to include in the rule. In addition, see EPAs response to comment EPA-HQ-OAR
2009-0923-1016-11 for a discussion that clarifies the component count requirements under
todays final rule and for discussion about the costs associated with the meters EPA intended to
include in the rule.
1055
Finally, EPA has required reporters to monitor emissions from above ground metering and
regulator city gate stations at which custody transfer occurs because these sources contribute
significantly to the total national emissions for the industry. See the Technical Support
Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923) for
complete discussion, including how EPA identified the major emissions sources and why those
emissions are subject to todays final rule. Furthermore, preamble III.E to todays final rule
explains the purposes the reporting of emissions under Subpart W.
Third, the infrared cameras are very expensive and not very plentiful. It will be difficult to
comply with this requirement regardless of whether consultants perform the work or we do the
survey in house. There are not enough cameras available or consultants capable of doing the
survey considering the number of existing M&R stations (270 city gate stations and over
500,000 M&R stations, depending on your definition). Even if EPA did not intend to include
residential meters in your definition of M&R, surveying even 270 city gate stations would be
time consuming and costly.
Response: EPA has revised todays final rule to allow alternative options. See preamble
Section II.F for details about the alternative leak detection options.
In addition, as described in EPA-HQ-OAR-2009-0923-1016-11, EPA has clarified that the rule
does not apply to customer meters, i.e., residential, commercial and industrial meters, served by
existing metering and regulating stations.
EPA reviewed its cost estimate and disagrees that surveying above ground city gate stations at
custody transfer would be overly time consuming and costly. Todays final rule allows reporters
to request the use of best available monitoring methods under certain conditions, such as such as
lack of cameras or consultants capable of doing leak surveys; please see the response to EPA
HQ-OAR-2009-0923-1011-27 for more information about use of best available monitoring
methods. As outlined above, todays final rule has clarified that leak detection is only required
at above-ground city gate stations with custody transfer and has revised todays final rule to
allow alternative methods of leak detection; therefore, far fewer resources are required to meet
the requirements of the rule than assumed by the comment. Furthermore, EPA is unable to
evaluate the commenters total estimate because insufficient documentation was provided to
explain how these figures were calculated. For example, the commenter did not specify their
assumptions about time required to perform leak detection at a metering and regulator station,
how many and what level staff members are necessary, what equipment must be purchased, or
what contractor costs would be incurred. Nonetheless, EPA has inferred that a large part of the
commenters cost estimate can be attributed to the misinterpretation that customer meters are
subject to leak detection requirements. The requirements do not apply to customer meters
(industrial, commercial, and residential meters) or to farm taps. See preamble Section II.F for a
complete discussion about the metersEPA intended to include in the rule. EPAs response to
1056

EPA intended to include in the rule.
Organization: Consumer Energy Company
Commenter: Amy Kapuga
EPA Has Vastly Underestimated Costs Changing or Eliminating the Annual Leak Survey
Proposal Would Help Reduce Compliance Costs Substantially
EPA is estimating a per company cost of $11,188 in the first year and about $7,000 per year per
LDC thereafter. In contrast, Consumers estimates the cost of complying with the leak survey
requirement in the rule will be more on the order of $1,000,000 per year, or more. EPA has
seriously underestimated the cost of complying with the 2010 Proposal, for at least four reasons.
* First, the 2010 Proposal uses the undefined terms above ground meter regulators in sections
98.232 and 98.233 118 , and metering stations and regulating stations in section 98.238 (defining
a natural gas distribution facility) that could be interpreted to include not only city gates and
large custody transfer or district metering and regulating (M&R) stations, but also industrial,
commercial and even residential customer regulating and metering equipment. We do not believe
this was EPAs intent, but unless the term is clarified, regulatory uncertainty and the risk of
varying interpretations by field enforcement personnel would drive LDCs to include all customer
meters in their leak surveys and reporting. When multiplied by each customer meter, the
significant annual leak survey costs would result in millions of dollars of unnecessary cost to
Consumers and our customers.
* Second, the agency has underestimated the cost of conducting an annual leak survey of the
eight listed components using optical scanning equipment at distribution city gates, above
ground meter regulator stations and at underground storage facilities. Requiring the use of
infrared cameras significantly increases this cost.
* Third, we do not have inventories or records for all eight of the components listed in section
98.232(f)(5) and (i)(1). We would have to visit each meter and regulator location (depending on
the definition of M&R) to develop the list required for applying component-level emission
factors to component counts. The burden of this proposal could be reduced significantly by
reducing the number of components and better defining them so it is clear what is to be included.
The burden could be eliminated by allowing the use of facility-level emission factors for city
gates and above ground M&R facilities, as EPA has proposed for below-ground M&R facilities
and vaults.
* Fourth, EPA has underestimated the costs making an initial threshold determination for small
distribution systems and other small facilities that likely fall below the threshold. These facilities
118
Proposed 40 C.F.R. 98.232.(i), 98.233(q)(7), 75 Fed. Reg. 18637, 18643.

1057
apparently would have to conduct a full leak survey using optical scanning in the first year in
order to determine whether a distribution, underground storage, transmission compression
facility or LNG facility does or does not exceed the 25,000 tpy regulatory threshold under
98.2(a)(2). EPA could avoid this burden by allowing small facilities to use a simpler threshold
determination method that does not require leak surveys or other field work.
1. Because M&R is Not Clear Defined and Could Apply to Customer Meters, Distribution M&R
Leak Survey Costs Could be Extremely Burdensome and Costly
Even without requiring infrared cameras, EPAs annual leak survey requirement would impose
far greater costs than it has estimated, largely because the proposal could be interpreted to apply
to customer meters. The cost of employee time for visiting each site is not large, but when
multiplied by each meters, the costs expand exponentially.
2. Optical Scanning/ Infrared Camera Costs
EPA explains in the preamble to the 2010 Proposal that EPA proposes conducting fugitive
emissions detection and then applying leaking component (or leak only) emissions factors for
processing, transmission, underground storage, LNG storage, LNG import and export terminals,
and LDC gate stations. 119 . Proposed section 98.233(q) requires LDCs, and other facilities
subject to Subpart W, to use the methods described in 98.234(a) to conduct an annual leak
detection of fugitive emissions from the relevant list of components in 98.232. 120
Consumers requests that EPA revise sections 98.232, 98.233 and 98.234 to clarify that the term
Natural Gas Distribution Facility does not include customer meters and that LDCs are not
required under Subpart W to leak survey or report fugitive emissions from industrial, commercial
and residential customer meters.
Response: EPA disagrees that it underestimated the rules cost for natural gas distribution
reporters. EPA determined that the commenter overestimated the cost based on assumptions that
were inconsistent with the intent of the rule. Response to comment EPA-HQ-OAR-2009-0923
1016-11, clarifies that monitoring requirements do not apply to customer meters, and that leak
detection requirements do not apply to all metering and regulator stations.
In addition, EPA did not intend to have reporters count the leaking components from all stations
and meters. See EPAs response to comment EPA-HQ-OAR-2009-0923-1016-11 for a
discussion that clarifies the component count requirements under todays final rule and why EPA
has elected not to use facility-level emissions factors for city gates and above ground M&R
facilities.
EPA agrees that screening tools would facilitate reporting determinations and plans to make such
tools available. Accordingly, EPA has updated the Economic Impact Analysis to better account
for reporting determinations and expected use of screening tools; see EPAs complete response
in Section III.B.2 of the preamble to todays final rule.
119
120
75 Fed. Reg. at 18622.

75 Fed. Reg. at 18642.
1058
Regarding the commenters concerns about leak detection equipment, see preamble II.F for
details about alternative leak detection options .

Organization: Latham & Watkins LLP on behalf of DCP Midstream

The High Cost and Extreme Burden Underscore The Lack of Justification for Super Source
Aggregation
EPAs estimate of the cost for reporting fugitive and vented emissions for all companies in all
segments of the petroleum natural gas industry covered by Subpart W is $56 million for first year
costs and $21.4 million for continuing annual costs. See Proposed Rule, Tables W-5 and W-6, at
p. 18625. As discussed below, DCP estimates that its first year costs alone, even without the
proposed super source aggregation, would be almost $63 million and thus exceed EPAs
estimate for the entire industry. Moreover, if EPA finalizes its proposed super source
aggregation, DCPs total capital costs to comply with the Rule would be over $200 million, or
almost four times EPAs estimate for the entire industry. Thus, EPA has significantly
underestimated the burden that would be placed on the processing sector were it required to
collect the information necessary to comply with the Proposed Rule.
A specific example of EPA having underestimated the costs of complying with the Proposed
Rule is the cost associated with the metering that the Proposed Rule would necessitate. [FN 26 Table 1 of these comments identifies the Proposed Rules Sections that, based on DCPs
interpretation, require the use of meters.] According to the Background Technical Support
Document, it appears EPA assumed a cost of $1,400 per meter for a simple pitot, annubar, or
rotometer. In many cases, however, these meters are not adequate devices for these applications.
Rather, DCP Midstream believes orifice plate meters and roots meters are necessary for these
metering requirements. Table 1 below provides DCPs estimates of its costs associated with
installing and operating the meters necessary to comply with the Proposed Rule. In addition to
the purchase price of the meter, DCP Midstream has factored in costs associated with
installation, outage time, straight pipe run requirements, and transmitting data to the companys
data collection system.
Under a non-aggregation source approach, DCP estimates that, at a minimum, most or all of its
48 processing plants and treaters and about 24 compressor stations, or 72 total sources, would
exceed the 25,000 tpy CO2e threshold when Subpart C emissions are combined with Subpart W
emissions. As detailed in the Table 1 below, DCP estimates its capital costs to monitor these
major sources at approximately $63 million. Under EPAs proposed super source
aggregation, an additional 630 of DCPs compressor stations, whose individual emissions fall
well below the 25,000 tpy CO2e threshold, would become subject to the reporting requirement.
DCP estimates that its total costs with the proposed super source aggregation would be about
1059
$203 million, or more than $167 million over and above its costs without such aggregation.
Table 1 of these comments identifies the Proposed Rules Sections that, based on DCPs
interpretation, require the use of meters.
Thus, the Proposed Rules super source aggregation would cost DCP alone an additional $167
million in first year capital costs.
In addition to the enormous capital costs, DCP estimates that its annual operating costs to
comply with the Proposed Rules super source aggregation would be approximately $6.5
million dollars, which is almost one third of EPAs estimate of $21.4 million for the entire
sector. The $6.5 million figure represents a more than three-fold increase in annual operating
costs over and above the $2 million per year DCP estimates it would cost to comply with the
Rule were EPA to apply the CAAs longstanding non-aggregation source approach.
Moreover, a disproportionate amount of DCPs additional costs to comply with the Proposed
Rules super source aggregation -- about $140 million -- would result from EPAs
misconception about the proportion of the sectors fugitive GHG emissions that originate from
compressors. This $140 million represents the difference in DCPs costs to determine emissions
and from the compressor rod vents and wet seals under a non-aggregation source approach
($59.3 million) and DCPs costs to monitor such units under a super source aggregation
approach ($199.3 million). As explained in more detail in Section VI.A of these comments, DCP
believes that EPAs estimate that 48% of the sectors fugitive emissions originate from
reciprocal compressors is greatly overstated. Indeed, information from another midstream
company indicates that such emissions may comprise less than 6% of the sectors total
emissions. This being the case, the additional $140 million it would cost DCP to determine
emissions from all its upstream compressor station sources under the proposed super source
aggregation approach cannot be justified.
1060
Response: With regard to super source aggregation, please see Topic 2: Aggregation of
Gathering and Boosting Systems with Processing Facilities in Volume 9 of the response to
comments to todays final rule (EPA-HQ-OAR-2009-0923). Todays final rule does not include
gathering system and boosting stations. Therefore, the commenters cost estimate for super
source aggregation is not relevant.
EPA disagrees with the commenters estimate for metering, which was based on the use of
higher-cost, optional methods. The commenter did not explain why they concluded that the
meters identified by EPAsimple pitot, annubar, or rotometersare inadequate for complying
with the rule. Although EPA considered various metering options, such as plate orifice meters
and root meters, EPA determined that lower cost options exist to measure emissions, such as port
installation and temporary meters. Therefore, EPA based its cost estimate on port installation
and temporary meters. The preamble to todays final rule clarifies that although the installation
of permanent meters are an option under the rule, temporary meters are acceptable to measure
emissions from reciprocating and centrifugal compressors. In addition, reporters are not required
to install a meter to determine compressor throughput; see preamble Section II.F for complete
discussion about the requirements. See preamble Section III.B for detailed discussion about the
costs associated with temporary meters and port installation.
Regarding acid gas removal units, EPA has revised the monitoring methods and provided several
options to estimate emissions that eliminate the need for new meters at these units. Please
response to EPA-HQ-OAR-2009-0923-1024-26 for further details on the flexible reporting
requirements for these units.
Regarding the 48 percent contribution of reciprocating rod packing emissions from the natural
gas processing segment, please see response to EPA-HQ-OAR-2009-0923-1154-6 for further
details.
Organization: Latham & Watkins LLP on behalf of DCP Midstream
Furthermore, EPA would appear, in large part, to rely on its overestimate of compressor fugitive
emissions to justify the enormous burden that its proposed super source aggregation would
place on the midstream sector. As explained in Section IV.C.3 of these comments, the
additional cost to DCP of testing vent packings in upstream compressors located outside of
major source facilities would be approximately $70 million. This cost alone is greater than
EPAs estimate of compliance costs for the entire sector and it cannot be justified in an attempt
to account for what is, at best, an inadequately documented proportion of fugitive emissions, and
that in fact may be less than 6% of total sector emissions.
Response: Please see the response to comment EPA-HQ-OAR-2009-0923-1154-10 for a
response to this comment.
1061

Organization: Laclede Gas Company
Commenter: Steve Donatiello
As has been emphasized throughout our comments, any type of gas leak, whether from rod
packings or any other system component, represents a potential safety issue that gas system
operators view seriously and address promptly. Volumetric leak testing on each of these units is
difficult at best, and in Lacledes view, is not a good use of already-stretched customer dollars
and labor resources.
Response: EPA disagrees that monitoring is an inappropriate use of resources EPA has required
reporters to monitor emissions from reciprocating compressor rod packing units because these
sources contribute significantly, about 28 percent, to the total reported emissions for the natural
gas and petroleum industry. See the Technical Support Document (TSD) for todays final rule
found in docket (EPA-HQ-OAR-2009-0923) for complete discussion, including how EPA
identified the major emissions sources and why those emissions are subject to todays final rule.
Finally, see EPA-HQ-OAR-2009-0923-1010-2 for EPAs response to comments about its efforts
to reduce the burden and costs of the rule, while still ensuring that the program yields high
quality data and essential information.
Organization: NiSource, Inc.
Commenter: Kelly Carmichael
It is difficult to assess the economic burden of collecting and reporting this information. EPA's
estimate for the burden on industry is orders of magnitude too low. For example, the cost of an
infrared camera ranges from $80,000 to $100,000 in today's market (prior to the publishing of
the final version of this rule by EPA). The cost for renting each of these cameras is about $2,500
per week. NiSource assumes that EPA definition of an M & R Station includes industrial
customer meters and regulators, but not commercial and residential meters and regulators. There
is no definition of "connectors" in the rule, therefore, we are unable to estimate the resources
required for cost estimation and compliance.
Response: EPA disagrees that its cost estimate is orders of magnitude too low. In line with the
commenter, EPA estimated a cost of $100,000 per camera. While EPA generally agrees with the
commenters assumption about the cost for an infrared camera, EPA disagrees with the
commenters interpretation about the extent of equipment subject to leak detection. See preamble
Section II.F for a complete discussion about the meters. EPA intended to include in the rule.
EPAs response to comment EPA-HQ-OAR-2009-0923-1016-11 discusses the costs associated
with the meters EPA intended to include in the rule.
1062
Furthermore, EPA disagrees that there is no definition of connectors in the rule; please see the
definition for connector as discussed in subpart A of The Final Mandatory GHG Reporting Rule
(40 CFR part 98).
Commenter: Jeff Applekamp
EPA has significantly underestimated the burden to industry to obtain the information to
calculate the emission estimate for gathering pipelines.
Response: EPA has removed this source from todays final rule. See Section II.F of the
preamble for further details.

EPAs estimated cost impact is at least an order of magnitude low.
GPA has evaluated the economic impact of the proposed Subpart W as compared with the results
to the analysis provided by EPA. Our cost analysis is provided in the following Table 1. Gas
Processing Plants and Table 2. Booster Stations with Table 3. a summary of assumptions made.
This analysis indicates that EPAs cost estimates are one to two orders of magnitude low. For
example, our analysis shows that the average cost of compliance for a gas plant for the initial
year will be about $2.4 million, as compared to EPAs estimate of $59,000. Similarly, our
analysis demonstrates that the average cost of compliance for a booster station for the initial year
will be about $298,000, versus EPAs estimate of $12,000 (a basis for a small gas plant since
EPA did not provide a cost breakout for a booster station).
It is clear that EPA has failed to recognize the full impact of requiring reporting at all facilities
and pipelines in the natural gas gathering and processing sector. Our analysis shows that EPA
has significantly understated the cost and burden of Subpart W as they are significantly higher
than EPA projected. GPA requests that EPA re-evaluate the costs associated with this rule, and
should significantly simplify and streamline reporting requirements for the midstream sector.
1063
1064
1065
1066
1067
1068
1069
Response: EPA disagrees that the April 2010 proposed rules total national cost estimates are
one to two orders of magnitude low for gas processing facilities; see below for more details.
EPA has decided not to include gathering lines and boosting stations as an emissions source in
todays final rule at this time, see Section II.F of the preamble for further details. As the
gathering and boosting costs estimated by this commenter account for $3.8 billion, or
approximately 85 percent of their total national cost, this clarification explains the majority of
the cost disparity between EPAs and the commenters cost models.
For natural gas processing, this commenter estimates a first year total national cost of $699
million compared with EPAs total national cost estimate for todays final rule of $7.8 million.
EPA has reviewed the assumptions embedded within this commenters cost model, and this
commenter made several assumptions that were not consistent with the proposed rules intent.
One was the assumption that EPA required the installation of permanent flow meters for
centrifugal and reciprocating compressors vents and compressors throughput, at a cost of
$100,000 per meter. This amounts to a total of $632 million. However, as discussed in Section
III.B.2 of the preamble to todays final rule, EPA does not require permanent meters for vents,
1070
and allows for installation of a port for using a temporary insertion flow meter for an annual
estimate of a one-time measurement of vented emissions from specified compressor vent
emissions. In addition, EPA is allowing in todays final rule the use of an acoustic detection
device (non-invasive) for detecting and quantifying through valve leakage for some of the
compressor emissions: closed blowdown vent valve leakage and closed compressor unit isolation
valve leakage through a depressurized compressor. EPA did not intend for reporters to install a
meter to determine compressor throughput. EPA has clarified this in todays final rule and allows
reporters to use engineering estimation to determine compressor throughput; see Section II.F of
the preamble for further details. EPA requires the compressor throughput for analysis of the
activity data and the resultant GHG emissions reports, as combustion CO2 will be proportional to
compressor throughput. Finally, the commenter assumed that the rule would cover all
compressors in natural gas processing, or about 5,552 compressors, which exceeds EPAs
estimated coverage of about 3,796 compressors at the 25,000 metric ton CO2e reporting
threshold. See Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background TSD (EPA-HQ-OAR-2009-0923) for EPAs detailed threshold analysis. These
factors explain the large disparity between the EPA and the commenters compliance cost, and
account for a difference in national cost of $627 million.
The commenter has also assumed for the natural gas processing segment the installation of
permanent flow meters to measure the acid gas removal vent stacks at a cost of $100,000 per
meter for a total national cost of $21.9 million. Please see preamble III.B.2 for EPAs response.
The commenter also assumed that in natural gas processing, quarterly gas sampling was required
for each compressor, dehydrator, and AGR unit for a total national cost of $9.8 million. As
explained in the Section II.F preamble of todays final rule, EPA modified the requirements by
replacing quarterly sampling of gas composition with the option of best available sample
analysis for compressors and dehydrator vents. AGR units require quarterly sampling, if CEMS
is not available or is not installed, or if a continuous gas analyzer is available or not installed.
This accounts for a difference in national cost of $7.3 million.
The commenter has assumed that to model emissions from storage tanks in the natural gas
processing segment, there would be costs associated with computer modeling, sampling and
analysis, and the installation of sampling ports for a total national cost of $3.9 million. EPA has
determined that natural gas processing facility storage tanks are not a significant emission
source. EPA has determined that most gas condensate coming into a processing facility will have
passed through a field or gathering atmospheric pressure storage tank and have been stabilized
(i.e. water has been removed and flash vapors have been released from the gas condensate).
Hence, the emissions from processing storage tanks are expected to be minimal. . Therefore, in
the todays final rule, EPA has provided a clarification by excluding the requirement for
monitoring natural gas processing storage tanks.
The commenter has assumed for natural gas processing data management system and reporting
and compliance management a total national cost of $7.0 million. Although the commenter did
not provide any information about the software represented in its analysis (except for cost), EPA
disagrees with the commenters cost estimate as even the largest of reporters under this final
action will be able to use or adapt from their data management and reporting obligations under
1071
subpart C the standard spreadsheets or databases to collect the emissions data and perform
calculations at a facility level. EPA accounted for data management costs by factoring in
estimates of labor to set up spreadsheets and other archiving and recordkeeping activities, as well
as equipment costs like file cabinets and external hard drives; see the EIA for a complete
discussion. EPA estimated the costs for data management and reporting under subpart W alone
to be $1.5 million in todays final rule. This accounts for a difference in national cost of $5.5
million.

Third, EPA has greatly underestimated the cost and effort to conduct the direct measurement
prescribed for these sources. Any one company will have hundreds or thousand of compressors
in operation, each one with multiple compressor rod packings. Additional complexity and cost
are presented in the many cases when a single compressor is used for multiple services, for
example inlet compression, residue compression and refrigeration compression.
Response: EPA disagrees that the proposed rule underestimated costs. This commenter
assumed that the rule would cover all compressors in natural gas processing, or about 5,552
compressors (see EPA-HQ-OAR-2009-0923-1206-25 for commenters detailed analysis). The
commenters estimate exceeds EPAs estimated coverage of about 3,796 compressors; please see
EPAs response to comment number EPA-HQ-OAR-2009-0923-1206-25.
On a related issue, Section 98.236(c)(18) requires reporting the gas throughput for each
compressor. The metering required for this reporting is not typically in place today and would
require installation of meters on each compressor at a cost of about $100,000 per meter. If a
company has 1000 compressors, the capital cost to collect this one piece of operational data is
$100 million for data this is not even necessary for calculating emissions from a compressor rod
packing.
Response: EPA disagrees with this comment, as the rule does not require the installation of
$100,000 permanent flow meters; see EPAs response to EPA-HQ-OAR-2009-0923-1206-25.
Organization: Northeast Gas Association
Commenter: Thomas M. Kiley
1072

This proposed rule will however add major costs against uncertain gains, given the uncertainty of
the methodology applied. Estimated compliance costs for natural gas utilities within these
regulatory parameters are high and raise serious concerns. The ability of a utility company to
recover these costs in a rate case is uncertain, particularly in a time of difficult economic
circumstances for most homeowners, businesses and average ratepayers. We would urge EPA to
consider the issue of compliance costs very carefully.
Response: EPA disagrees that the proposed rules compliance costs were underestimated. For
EPAs response to comments that it underestimated costs for the distribution segment, see
comment EPA-HQ-OAR-2009-0923-1152-4. EPA has determined that the burden to LDCs is
minimal and did not incorporate any assumptions in the analysis about passing costs on to
consumers in the form of higher rates. However, even if reporters pass on the cost to their
customers the effect on product prices will be insignificant. Please see EPA-HQ-OAR-2009
0923-1303-2, for EPAs response to the comment regarding compliance costs and the passing on
of these costs to consumers.
EPA has thoroughly reviewed and updated its cost estimates to reflect changes in todays final
rule (e.g. incorporating the capital cost to purchase a laser emissions detector versus an optical
imaging device). See Section II.E and II. F of the preamble for major changes to the rule and
Section III of the preamble for a summary of this analysis and the Economic Impact Analysis for
a detailed discussion on costs.

Organization: DTE Energy
Commenter: Gregory L. Ryan
The costs to implement Subpart W for LDCs have been grossly underestimated by EPA and are
not indicative of a reasonable cost to measure the small amount of GHGs emitted from natural
gas LDCs.
EPA has estimated that the total industry-wide annual cost of complying with Subpart W for
natural gas distribution will be $1.6 million in the first year and $1 million per year thereafter. 121
EPA estimates that 143 LDCs will be subject to Subpart W using a facility threshold of 25,000
tons per year CO2e. 122 Accordingly, EPA is estimating a per company cost of $11,188 in the first
year and about $7,000 per year per LDC thereafter. These costs are grossly underestimated for
the anticipated level of work that would be required to be performed on the MichCon distribution
system for DTE Energy to comply with Subpart W.
121
75 Fed. Reg. at 18625, Tables W-5 and W-6.
122
EPA, Economic Impact Analysis for the Mandatory Reporting of Greenhouse Gas Emissions Under Subpart W
Supplemental Rule (GHG Reporting), Final Report (Dec. 8, 2009) (Economic Analysis) at page 4-5.
1073
DTE Energy (MichCon) owns and operates 467 city gates/custody transfer stations,
approximately 600 above grade district regulators, and approximately 1,700 below grade district
regulators. DTE Energy also employs a significant number of farm taps and industrial meter
and regulator (M&R) stations which serve to lower the natural gas pressure to the level required
to provide gas to the end user. As currently written, DTE Energy believes that the proposed rule
would require all 467 city gates/custody transfer stations, 600 above grade district regulators, all
farm taps, and potentially all the industrial M&R stations to perform leak detection by optical
imaging equipment. These metering and regulating stations are located throughout the state of
Michigan, including the Upper Peninsula.
Assuming that DTE Energy utilizes its own internal resources for conducting the required
monitoring of the MichCon LDC, we anticipate the first year costs to exceed $100,000 alone for
purchase of optical imaging equipment and training of personnel on the equipment. Furthermore,
we estimate that we will need at least two full time equivalent (FTE) employees to conduct the
required annual leak detection monitoring for the MichCon distribution system. This estimated
level of effort includes the time to conduct leak tests at more than 1,000 locations, driving time,
and data management activities required to inventory, record, and update leak records from
thousands of components. The level of effort to comply with the rule for LDCs has been grossly
underestimated by EPA and is not appropriate for measuring GHG emissions that account for
less than 1% of total U.S. GHG emissions.
Response: EPA disagrees that the proposed rules costs were grossly underestimated. First, the
requirements do not apply to customer meters (industrial, commercial, and residential meters) or
to farm taps. See preamble Section II.F for a complete discussion about the meters EPA
intended to include in the rule. EPAs response to comment EPA-HQ-OAR-2009-0923-1016-11
discusses the costs associated with the meters EPA intended to include in the rule.
Similar to the commenters analysis, EPA assumed reporters would need to visit M&R stations,
but unlike the commenter, did not estimate costs for visits to customer meters because they are
not subject to the rule. Please see EPA-HQ-OAR-2009-0923-1016-11 for further details on
EPAs assumptions about component count requirements and discussion about sources covered
under the rule.
EPA is unable to evaluate the commenters estimate of $100,000 for optical leak detection
equipment and training because insufficient documentation was provided to explain how this
figure was calculated. For example, the commenter did not specify their assumptions about the
number of cameras required to monitor their sources, the cost of a camera, nor the duration and
cost of training assumed. Also, EPA has annualized all costs over the life of the equipment (5
years in the case of an optical camera) and at a discount rate of 7 percent per annum. Finally,
EPA has revised todays final rule to allow alternative options for leak detection. See preamble
Section II.F for details about the alternative leak detection options, such as the use of Method 21
to conduct leak detection surveys. This new option may potentially offer a less costly alternative
for the commenter. See EPAs response to EPA-HQ-OAR-2009-0923-1020-6 for details about
EPAs assumptions for the costs associated with optical leak detection equipment.
1074
Furthermore, the commenter did not explain why at least two full time equivalent employees
would be required to comply with the rule. See EPA-HQ-OAR-2009-0923-1016-11 for EPAs
response to concerns about the costs for leak detection; see preamble Section II.F for details
about permitted alternative leak detection options.

Local Distribution Companies (LDCs) Should Not be Required to Report Under the Mandatory
Greenhouse Gas Reporting Rule
According to EPAs Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008,
emissions of methane (CH4) and non-combustion CO2 from natural gas distribution systems
account for less than 1% of U.S. total CO2e emissions. The level of effort to measure natural gas
local distribution company (LDC) emissions that is proposed in the rule is not commensurate
with the magnitude of the expected emissions from natural gas LDCs. This level of effort
translates into significant added resource costs for monitoring equipment (e.g. leak detection
instrumentation), data-base management, training, and the cost of technicians to visit, inspect,
log, and report fugitive emissions from components on a natural gas distribution system.
Response: As described in EPAs response to EPA-HQ-OAR-2009-0923-0837-9, EPA
disagrees that it underestimated the compliance costs. Regarding the commenters concerns
about how the level of effort to monitor emissions compares to the value of the data, see EPAs
The costs to implement Subpart W for LDCs have been grossly underestimated by EPA and are
not indicative of a reasonable cost to measure the small amount of GHGs emitted from natural
gas LDCs The proposed rule requires significant additional effort to log and report leaks beyond
current requirements for detecting and fixing leaks on distribution systems, and does not
recognize that existing programs already minimize fugitive emissions from LDCs
Response: As described in the response to comment EPA-HQ-OAR-2009-0923-1152-4 above,
EPA disagrees that the proposed rules compliance costs were underestimated. For response to
the commenters concerns about how current leak detection practices relate to todays final rule,
see comment EPA-HQ-OAR-2009-0923-1156-14.
10.4
ECONOMIC IMPACTS
1075

Executive Order 12866
Finally, EPA neglected to include the full economic impacts of proposed Subpart W in the
analysis required by EO 12866. EO 12866 directs EPA to submit to OMB new significant
regulations under consideration by the Agency. 123 EPA states that it submitted proposed Subpart
W to OMB for review under Executive Order 12866. However, the Agency failed to include
anything close to the full costs of the rule in the Economic Impact Analysis and there is no
indication that EPA included these costs in its submission to OMB. Without this key
information, OMB could not fully review the impacts of the proposed rulemaking.
Response: EPA disagrees with this comment. As stated in the preamble Section IV and in the
Economic Impact Analysis, EPA determined that this rulemaking is a significant regulatory
action pursuant to Executive Order 12866 because it raises novel legal or policy issues arising
out of legal mandates, the Presidents priorities, or the principles set forth in EO 12866.
Accordingly, EPA submitted the Economic Impact Analysis to the Office of Management and
Budget (OMB) for review under Executive Order 12866 as part of the proposed and final
rulemakings, thereby meeting all the requirements of this Executive Order.
EPA relied on best available data and methods to estimate the costs and economic impacts of the
rulemaking and documented the analysis and results in the Economic Impact Analysis, thereby
facilitating a complete OMB review. As discussed in the preamble Section III, EPA collected
and evaluated cost data from multiple sources, thoroughly reviewed the input received through
public comments, and weighed the analysis prepared for the proposal against this input. In
finalizing the rule, EPA has determined that the Economic Impact Analysis provides a
reasonable characterization of costs and that the documentation provides adequate explanation of
how the costs were estimated.
For detailed responses to the cost and economic comments from this commenter, see EPA-HQ
OAR-2009-0923-1151-89 through EPA-HQ-OAR-2009-0923-1151-107 for further details.
If the required sources to report are aggregated by basin, the increased number of facilities
requiring travel to remote areas has not been taken into consideration. The EPAs cost
estimations also assume that staff exists on-site to assist with GHG reporting implementation: the
123
Executive Order 12866 Section 6(a).

1076
opposite is true. Many operators will be forced to hire additional temporary staff or utilize
consultants to review and compile data from various divisions in order to determine if a company
is in compliance with the proposed rule. Discussions with state regulators on this very point has
demonstrated that a new cottage industry of greenhouse emissions auditors will need to be found
and adequately trained to assist the hundreds of small businesses in need of reporting data. In the
preamble to this rule, EPA notes that it completely changed the required methodologies and now
requires more reporting from smaller sources that have limited or no emissions data. Direct
measurement is required for transmission tanks, wet seal degassing, reciprocating rod packing
venting, flare stacks. Requiring direct measurement of transmission tanks, wet seal degassing
and reciprocating rod packing venting will require field personnel to visit each site to determine
which sources must be reported, and to measure each of those sources. The direct measurement
of the small sources is costly, time consuming and will require significant use of consulting
sources to achieve compliance with the proposed rule.
Response: EPA disagrees that operators will need to hire additional temporary staff or use
consultants to determine whether or not they have to comply with Subpart W; EPA also
disagrees with the commenters assumptions about travel costs. Please see EPAs response to
EPA-HQ-OAR-2009-1060-1 and EPA-HQ-OAR-2009-0923-1151-89 for the complete response.
Furthermore, EPA determined that screening tools would facilitate reporting determinations and
plans to make such tools available. Accordingly, EPA has updated the Economic Impact
Analysis to better account for reporting determinations and expected use of screening tools; see
EPAs complete response in Section III.B.2 of the preamble to todays final rule.
This commenter also expressed concerns about the impact of the rule on small businesses. See
EPA-HQ-OAR-2009-0923-1005-7 for EPAs response, which includes a discussion about the
analysis supporting EPAs conclusion that the rule will not have a significant economic impact
on a substantial number of small entities.
EPA agrees with the comment that the direct measurement of emissions from compressors and
storage tanks in the onshore production sector is overly burdensome for the emissions covered.
Please refer to Section II.E of the preamble (EPA-HQ-OAR-2009-0923) for changes EPA has
made to the Rule with respect to storage tanks and reciprocating compressors.
However, EPA determined the direct measurement of emissions from condensate storage tanks,
compressors, and flare stacks in natural gas transmission sector is necessary in order to obtain
data of sufficient quality to inform policy. Please see the response to comment EMAIL-0002-9
OAR-2009-0923) for complete discussion about why EPA selected direct measurement for these
sources.
Organization: Texas Department of Agriculture
Commenter: Todd Staples
Placing additional regulatory and reporting burdens on this industry will certainly increase the
1077
cost of both agricultural production and consumer consumption of food and fiber.
Inelasticity and susceptibility to price volatility are defining characteristics of the agricultural
marketplace. As the industries providing the basic inputs agriculture relies on (fertilizers,
chemicals and fuels in this case), see increased costs due to the burden of greenhouse gas
emission reporting, so will agriculture producers, and therefore all American consumers.
According to the USDA, 13 percent of agricultural producers' costs of production result from
these input costs. Unlike many other industries, however, agriculture has few options in curbing
its demand for these inputs and only one option for passing along cost increases to third partiesto reduce supply.
Likewise, consumers are unable to curb demand for our basic life necessities-food and fiber. The
resulting impact of this economic squeeze will likely be reduced food and fiber supply and
increased consumer prices.
Response: EPA does not agree with the comment. EPA would like to reiterate that todays final
rule collects information related to GHG emissions from the petroleum and natural gas industry
to inform future policy; the rule does not regulate GHG emissions. See EPA-HQ-OAR-2009
0923-1010-2 for EPAs response to comments about its efforts to reduce the burden and costs of
the rule, while still ensuring that the program yields high quality data and essential information.
Also, EPA disagrees with the commenters conclusion that todays final rule will increase the
price of fuels. Although the commenter did not provide documentation or information to
substantiate the conclusions, EPA has nonetheless conducted an analysis to consider the potential
impact of price increases. EPA determined that even if reporters were to pass all costs associated
with todays final rule to consumers in the form of higher petroleum and natural gas prices, the
marginal price change would be minimal. In sum, EPA estimates that this price increase would
be less than one cent per barrel of oil or thousand cubic feet of gas, which is about 0.1 percent of
2006 oil and gas prices. (Please refer to the Subpart W Greater Economic Impact docket memo
EPA-HQ-OAR-2009-0923). EPA considers that such a small burden is justified to collect
necessary information that will help formulate cost-effective policies in the future. This analysis
remains consistent with EPAs conclusion that the potential benefits of more comprehensive
information about GHG emissions outweigh the costs of todays final rule.
10.5
IMPACTS FOR SMALL ENTITIES

1078

The commenter is concerned about the expense of reporting for production. Basin level reporting
would likely require reporting from small or marginal wells that might otherwise fall below a
field or well level threshold. There are numerous marginal wells in existence, and this will have a
large economic consequence to the profitability of each well. The environmental cost is additive
to the well development cost, which may result in plugged and abandoned wells.
Response: EPA recognizes the commenters concerns about costs associated with todays final
rule and has made every effort to reduce the burden and costs of the rule, while still ensuring that
the program yields high quality data and essential information. As discussed in Section II of the
preamble, todays final rule includes alternative data collection methodologies to target major
emitting equipment sources. While todays final rule will require reporting from some marginal
wells, EPA has determined that the associated burden will be minimal. First, todays final rule
requires reporters to inventory only major equipment to estimate emissions, not every well nor
all equipment at a well. Please see response to EPA-HQ-OAR-2009-0923-1151-5 for discussion
about changes EPA made to todays final rule to reduce burden for onshore production, such as
requiring operators to count their major equipment instead of each individual component.
Reporters will not need inventory most marginal wells because these wells typically do not have
major equipment due to the small volumes of production. In other cases, multiple marginal wells
will be connected to the same equipment; an operator would count the equipment but would not
need to do an inventory at the attached marginal wells. Second, todays final rule has several
equipment thresholds that in most cases result in insignificant burden on marginal wells; EPA
expects most marginal wells to fall below the equipment threshold, thereby allowing the reporter
to use an emissions factor based approach. See also EPA-HQ-OAR-2009-0923-1005-7, which
summarizes EPAs analysis of todays final rules economic impact on small entities and the
basis for EPAs conclusion that todays final rule does not target marginal operators who own
few wells and have small volumes of production.
While EPA notes that basin-level reporting may require reporting from some marginal wells,
EPA has concluded that basin-level reporting will more directly and completely meet EPAs
objective to collect accurate and comprehensive data on emissions across the petroleum and
natural gas sector to better inform future public policy. In particular, EPA analyzed emissions
coverage at different thresholds for the field- and basin-level options and found that basin-level
reporting is the more cost-effective option. See the Chapter 5 of the Economic Impact Analysis
for discussion of the analysis and results (EPA-HQ-OAR-2009-0923).
EPA agrees that the compliance costsi.e., the environmental cost noted by the commenter
will be additive to well development cost. EPA has therefore selected the most cost effective
reporting threshold and facility definition. Please see Section II.C of the preamble to the April
2010 proposal (EPA-HQ-OAR-2009-0923) for more details on the rationale for choosing basinlevel reporting over field- and well-level reporting.
1079
Small Business Implications

EPA cavalierly asserts that this proposal will not have a significant economic impact on a
substantial number of small entities. But, can this be true? Comparing numbers of wells that
must report against the number of wells operated by small businesses shows a different result.
In creating its basin-level reporting approach, EPA indicates that it will capture 81 percent of the
onshore petroleum and natural gas production GHG emissions. It also states in rejecting the
logical well pad facility definition that individual well pad emissions were low. Consequently,
we must conclude that EPAs definition must capture something close to 80 percent of the
operating wells.
In 2008, there were 960,303 operating wells in the U.S. (525,287 oil wells and 435,016 natural
gas wells, with about 7,000 of these in the federal offshore). The Energy Information
Administration reports that 85 percent of these oil wells and 73.3 percent of these natural gas
wells are marginal wells. Assuming a proportional distribution across wells, the following results
would be produced:
Clearly, there will be a pervasive burden borne by Americas marginal well producers. EPA is
well aware that the companies operating marginal wells are dominated by small businesses. To
suggest that the proposed rule will not have a significant impact on small businesses is simply
incorrect.
Response: EPA analyzed the economic impact of todays final rule on small entities in
accordance with the Regulatory Flexibility Act and the Small Business Regulatory Enforcement
and Fairness Act. This analysis revealed that the average ratio of annualized reporting program
costs to receipts of establishments owned by model small enterprises was less than 1 percent for
industries presumed likely to have small businesses covered by the reporting program, except for
entities in onshore production with 1-20 employees, for which the ratio was between 1 and 2
percent. Although a majority of the enterprises in the petroleum and natural gas industry fall in
the 1-20 employee range, most of the production, and therefore emissions, comes from large
corporations. In fact, the top 20 large corporations in the U.S. account for 85 percent of the total
oil and 77 percent of the total gas produced in the country. Small enterprises have very small
operations (such as a single family owning a few production wells) that are highly unlikely to
cross the 25,000 metric tons CO2e reporting threshold.
In addition, EPA disagrees with the commenters analysis of the number of wells subject to
reporting; the percentage of GHG emissions covered under the rule does not equal the percentage
of wells subject to reporting. Please see EPAs response to EPA-HQ-OAR-2009-0923-1151-89
for a discussion regarding the number of affected wells and the associated costs.
1080
Furthermore, the operation of marginal wells is not limited to marginal operators; large operators
own marginal wells. EPAs threshold analysis shows that todays final rule does not target
marginal operators who own few wells and have small volumes of production. Using the HPDI
database, EPA estimated that of the 19,876 marginal operators in the United States, only 2
percent (or 349 marginal operators) would meet the reporting threshold. (Marginal operators are
facilities with an average petroleum production below 10 barrels per day per well and an average
natural gas production below 75 thousand cubic feet per day per well.) Most marginal reporters
do not typically operate equipment that exceeds the equipment thresholds for separators and
dehydrators. In addition, any compressors at the marginal wellhead will use population
emissions factors. Therefore, marginal operators that meet the reporting threshold would likely
be able to report emissions primarily using population factors. See EPA-HQ-OAR-2009-0923
0053-1 for further discussion about final rule revisions that minimize burden associated with
reporting from marginal wells.
Based on this analysis, EPA concluded that this action will not have a significant economic
impact on a substantial number of small entities. The summary of the factual basis for the
certification is provided in Section IV of the preamble for the rule. Complete documentation of
the analysis can be found in the final Economic Impact Analysis (EIA), Section 5.2 (EPA-HQ
OAR-2009-0923). For further response to comments on owners of marginal wells, please see
EPA-HQ-OAR-2009-0923-0053-1.
Finally, EPA determined that screening tools would facilitate reporting determinations and plans
to make such tools available. In particular, the screening tools will assist marginal operators and
other entities in determining whether or not they meet the threshold. See EPAs complete
response in Section III.B.2 of the preamble to todays final rule.
In sum, this analysis is consistent with EPAs conclusion that todays final rule will not have a
significant economic impact on a substantial number of marginal U.S. operators. Please see
EPAs response to EPA-HQ-OAR-2009-0923-0053 -1 for further discussion about Subpart Ws
effect on marginal wells.
Impacts to small crude oil and natural gas businesses will be significant. Many oil and gas
operators in Oklahoma are small businesses (similar to small family farms). These small
independent oil and gas companies are ill-equipped to bear the costly burdens of GHG reporting
requirements. The risk tolerance of these small independent producers is significantly different,
and significantly less, as compared to large, integrated oil companies. In addition, most of these
smaller oil and gas businesses operate marginal wells. Marginal wells are mature crude oil and
natural gas producing properties that have lost their initial, high production rates and instead
operate on the much lower, flat end of the natural production decline curve. Each well has its
own unique economic hurdles that dictate whether the well is produced or not. The draft rule will
require oil and gas companies to conduct very detailed inventories of all equipment that could
emit GHGs even down to counting all threaded connections! These inventories would then be
1081
required to be evaluated in detail to determine their leak potential and only then could the
operator calculate the CO2e emission rate. It is clear that small operators would have to hire
consultants to do this work for them which would undoubtedly force many wells to be plugged
due to their then uneconomic status. Our country will lose the much needed domestically
produced energy and will lead to many small businesses failing.
The Interstate Oil and Gas Compact Commission define marginal wells as those that produce 10
barrels of oil per day or less, and less than 60 thousand cubic feet per day. In Oklahoma,
marginal wells produce on the average 1.7 barrels of oil per day and 24.3 thousand cubic feet of
gas per day. The emissions from these types of facilities are low, so low they are not required to
have an air emission permit. Despite the low production rates and pressures of marginal wells,
approximately 19 percent of the U.S. oil production and 8 percent of the natural gas produced in
the lower 48 states comes from marginal wells. EPA must carefully analyze the impacts to
marginal wells and ensure that any increased regulatory costs are clearly justified, especially in
these difficult economic times.
We believe that EPA has underestimated the number of operators and wells impacted and the
cost impacts on small crude oil and natural gas operators of marginal wells located in our state
and around the country. We believe this is fundamentally inconsistent with efforts to protect and
enhance national security by increasing domestic energy production and reducing dependency of
foreign oil. We do not believe EPA has fully addressed Executive Order 13211, and request EPA
reanalyze this issue.
Response: EPA understands the commenters concerns and has made every effort to reduce the
burden and costs of the rule, while still ensuring that the program yields high quality data and
essential information. EPA also recognizes the importance of considering the rules impacts on
small entities and has analyzed the potential for such impacts; see EPA-HQ-OAR-2009-0923
1005-7, which summarizes EPAs analysis of todays final rules economic impact on small
entities and the basis for EPAs conclusion that todays final rule does not target marginal
operators who own few wells and have small volumes of production. Furthermore, while todays
final rule will require reporting from some marginal wells, the associated burden will be
minimal. See EPA-HQ-OAR-2009-0923-0053-1 for further discussion about final rule revisions
that minimize burden associated with reporting from marginal wells.
EPA determined that screening tools would facilitate reporting determinations and plans to make
such tools available. In particular, the screening tools will assist small entities in determining
whether or not they meet the threshold. EPA expects the screening tools will assist most small
entities in their threshold determination.
Overall, this analysis is consistent with EPAs conclusion that todays final rule will not have a
significant economic impact on a substantial number of marginal U.S. operators.
EPA disagrees that it did not fully address EO 13211, Actions Concerning Regulations That
Significantly Affect Energy Supply, Distribution, or Use, (66 FR 28355 (May 22, 2001)). This
EO requires federal agencies undertaking certain regulatory actions to prepare a Statement of
Energy Effects that describes the adverse effects of a significant energy action on energy
supply, distribution and use, reasonable alternatives to the action, and the expected effects of the
1082
alternatives on energy supply, distribution and use. EO 13211 applies only to regulatory actions
leading to significant adverse energy effects. A significant adverse energy effect would occur
under this action if todays final rule were to increase the costs of energy production or cost of
energy distribution by one percent.
EPAs analysis, Subpart W Greater Economic Impact, (EPA-HQ-OAR-2009-0923),
demonstrates that the rule will not have a significant, adverse impact on the supply
(productivity), competition, or prices in the energy sector. Specifically, the analysis revealed
that that even if all of the costs associated with complying with todays final rule were passed on
to consumers in the form of higher petroleum and natural gas prices, the marginal price change
would be minimal. EPA estimates that the price increase would be less than 0.1 percent increase
in natural gas price and a 0.01 percent increase in crude oil price. This is significantly lower than
the one percent increase in fuel price (assuming a price of $7.70 per Mcf of natural gas 124 and
$58.15 per Barrel of oil 125 ) criterion as it applies to EO 13211. Given that the impact of the rule
on fuel prices is minimal, EPA does not expect the rule to affect demand for fuels. Therefore,
the rule will not affect current and future production from operators who build production
capacity based on perceived demand. In addition, the rule does not require the operators to shut
down either parts or entire operations for monitoring emissions sources that could lead to supply
disruption.
In sum, EPA concluded that this rule would not have a significant, adverse impact on the supply
(productivity), competition, or prices in the energy sector. As a result, EPA has determined that
EO 13211 does not apply to todays rule because it does not have any significant adverse impact
on energy supply, distribution, or use.
Basin level reporting is not appropriate for onshore crude oil and natural gas production sites.
EPA proposes that emissions from onshore crude oil and natural gas production sites be reported
at the basin level. As previously stated, defining a "facility" to include many wells over a very
large area is contrary to the Clean Air Act definition, and will subject many small businesses to
the proposed GHG data collection and reporting requirements that is directly opposite of EPA's
stated goal to limit the impacts to small businesses. We believe the proposed reporting rule will
be unreasonably burdensome and costly. Operators hiring consultants will incur numerous
mobilization fees as the consultants go from site to site across a basin. This will no doubt
negatively impact most wells, especially marginal wells, such that many may become
uneconomic to produce. In addition, EPA requests comments on reducing the reporting threshold
124
EIA. (2008) Annual Energy Review. Table 6.7. Retrieved on July 28, 2010 at
http://www.eia.doe.gov/emeu/aer/pdf/pages/sec6_17.pdf. A 5-year average from 2003 to 2008.
125
EIA. (2008) Annual Energy Review. Table 5.18. Retrieved on July 28, 2010 at
http://www.eia.doe.gov/emeu/aer/pdf/pages/sec5_45.pdf. A 5-year average from 2003 to 2008.
1083
from 25,000 mtCO2e to 10,000 mtCO2e for onshore petroleum and natural gas production. This
would be even more problematic for all operators, but especially for small businesses. EPA has
the flexibility in determining who should be subject to the reporting requirements. We request
EPA not use a basin level approach that will unnecessarily include many operators of wells
(including marginal wells) in the reporting requirements for onshore petroleum and natural gas
production industry.
Response: Please see comment EPA-HQ-OAR-2009-0923-1010-1 for a response to the
comment on the definition of a facility. For EPAs response to comments on the rules
impacts on small entities and on operators of marginal wells, see EPAs response to comment
EPA-HQ-OAR-2009-0923-1005-7. Please see the response to EPA-HQ-OAR-2009-0923-1151
89 for a discussion regarding the number of affected wells and the subsequent burden. Please see
EPAs response to EPA-HQ-OAR-2009-0923-1196-7 for discussion about why EPA disagrees
that operators will need to hire contractors to comply with Subpart W.
Indeed, USEPA acknowledges in the Preamble to its proposal that its novel basin-level
aggregation approach was adopted precisely because individual production wells generally will
fall below the very threshold that USEPA selected, after careful evaluation and analysis, as
representing the most appropriate balance between the amount of GHG emissions to be reported
and the burden on small emitters. See 75 Fed. Reg. at 18615. As intended, the basin-level
approach will require operators to report emissions from small or marginal wells that emit GHGs
in quantities far below 25,000 tpy CO2e 126 Not surprisingly, the burden associated with the
proposed basin-level approach will be considerable, as the costs associated with monitoring and
reporting the emissions from these marginal wells will add to the existing development and
operational costs, resulting in substantial economic consequences for the profitability of such
wells. Indeed, IOGA-WV harbors serious concerns that the significant costs associated with
conducting this monitoring---even if only for purposes of determining Subpart W's applicabilitywill result in the premature capping, plugging and abandonment of these small or marginal wells,
despite the fact that they remain viable energy resources that should be a vital part of this
country's movement towards "green" energy development. Further, the economic impact of
abandoning these wells will extend far beyond the individual well owners to employees and
contractors of those companies, their families, affected mineral owners, and the broader local
communities, which will lose tax revenues and other benefits associated with the continued
operations of these wells-all consequences that should be avoided in the nation's still-struggling
economy. Such outcomes directly contradict USEPA's stated goal to "balance the rule coverage
to maximize the amount of emissions reported while excluding small emitters. 74 Fed. Reg.
16448, 16456 (April 10,2009).
126
IOGA-WV refers USEPA to the statistics set forth in EPAs comments, which represent that approximately 85%
of oil wells and 74% of natural gas wells are marginal wells producing less than 15 barrels/day of oil and 90
mcf/day of natural gas, respectively.
1084
Response: For EPAs response to comments on the rules impacts on small entities and on the
operators of marginal wells, see EPAs response to comment EPA-HQ-OAR-2009-0923-1005-7.
Please see the response to EPA-HQ-OAR-2009-0923-1151-89 for a discussion regarding the
number of affected wells and the subsequent burden. Finally, see the memo on Subpart W
Greater Economic Impact (EPA-HQ-OAR-2009-0923) for discussion about the Final
Greenhouse Gas Reporting Rules impact on the greater U.S. economy and its citizens.

Organization: Western Business Roundtable
Commenter: Holly Propst
EPAs Approach Would Cause Significant Burden on Small, Independent Producers
We are particularly concerned about the potential impact of this rule on onshore oil and gas
producers in the West. The vast majority of the onshore facilities that would be forced to report
under the rule are marginal wells. Entities operating such marginal wells -- particularly in our
region -- are predominately small businesses. The regulatory burden on such enterprises would
be significant. EPA seems to inappropriately minimize these impacts in its analysis.
Response: For EPAs response to comments on the rules impacts on small entities and the
burden for operators of marginal wells, see EPAs response to comment EPA-HQ-OAR-2009
0923-1005-7 and EPA-HQ-OAR-2009-0923-0053-1.
EPA makes the following conclusion in the preamble to the proposed Subpart W.
[The rule would] requir[e] only a small fraction of total facilities to report. 75 Fed. Reg. at
18619.
The proposed Subpart W in fact increases coverage of gas gathering and processing facilities to
nearly 100% regardless of facility size, which is unduly burdensome and neither reasonable nor
appropriate.
Response: The statement identified by the commenter described the fraction of entities from all
covered segments subject to reporting, not the coverage of entities from a particular segment.
Under todays final rule, EPA expects a small fraction of entities in the domestic oil and gas
industry to be subject to reporting; see Section 4.4 of the Economic Impact Analysis (EIA)
1085
(EPA-HQ-OAR-2009-0923) for complete details on the number and share of entities and
emissions covered. As discussed in the EIA, EPA analyzed various reporting thresholds to
identify one that would ensure maximum emissions reporting coverage with minimal burden on
the industry
Regarding the commenters conclusion that 100 percent of entities in the gas gathering and
processing segment would be covered, todays final rule does not include gathering lines and
boosting stations as an emissions source in subpart W at this time, see comment response EPA
HQ-OAR-2009-0923-1206-13 and Section II.F of the preamble for full details. Of the remaining
processing facilities that could be candidates to report based on source definition, only 289 or 51
percent of all processing facilities are expected to meet or exceed the 25,000 Mt CO2e threshold.
As such, EPA disagrees with the commenters conclusion that nearly 100 percent of gathering
and processing facilities must report.

Organization: Linn Energy
Commenter: Paul M. Espenan
LINN Energy believes that EPA has not matched the rigor and cost of proposed emissionestimating methods to potential emissions. As a result, the regulatory burden and associated costs
to operators, in particular smaller companies, of stripper wells, is disproportionate to the benefit.
As proposed, EPA makes no adjustment to estimation methods so that the amount of effort is
appropriate given the potential emissions.
Response: EPA is unable to evaluate the comment regarding the rigor of the Agencys cost
estimates because the commenter did not provide any specific information to explain their
conclusion. Please see EPA-HQ-OAR-2009-0923-1005-7 for EPAs response to comments on
the rules impacts on small entities; EPA-HQ-OAR-2009-0923-0053-1 for EPAs response to
comments on rules impacts on operators of marginal wells; EPA-HQ-OAR-2009-0923-1010-2
and EPA-HQ-OAR-2009-0923-1080-49 for EPAs response to comments regarding the rules
costs and steps the agency took to minimize the burden while ensuring the program will yield
high quality data.
Regulatory Flexibility Act (RFA)
The RFA requires an agency to prepare a regulatory flexibility analysis of any rule subject to
notice and comment rulemaking requirements under the Administrative Procedure Act or any
1086
other statute unless the agency certifies that the rule will not have a significant economic impact
on a substantial number of small entities.127 A small entity is defined as a small business, small
organization and/or a small governmental jurisdiction, which include many small owners and
operators of onshore and offshore petroleum and natural gas facilities. 128 EPA failed to conduct a
regulatory flexibility analysis of the proposed Subpart W, because it certifi[es] that this action
will not have a significant economic impact on a substantial number of small entities. 129
However, EPA dramatically undercounted the burdens to small businesses who own and/or
operate petroleum and natural gas systems, who will be dramatically impacted by the rules
unprecedented monitoring, recordkeeping, and reporting requirements for small entities. EPA is
obligated to at least repeat its analysis using more accurate economic impact data in order to
truly determine whether the proposed rule impacts small businesses.
Response: EPA disagrees with this comment. First, EPA analyzed the economic impact on
small entities using the revised cost estimates discussed in preamble Section III and in the EIA.
These cost estimates reflected improvements made in response to comments as well as changes
to the monitoring requirements in todays final rule. See preamble III.D for further discussion
about EPAs compliance with the Regulatory Flexibility Act; Section IV of the preamble and
Sections 5.2 and 6.3 of the Economic Impact Analysis document in detail the analysis EPA
conducted in accordance with the Regulatory Flexibility Act. .
In addition, EPA determined that screening tools would facilitate reporting determinations and
plans to make such tools available. In particular, the screening tools will assist small entities in
determining whether or not they meet the threshold. EPA expects the screening tools will assist
most marginal operators in their threshold determination. See EPAs complete response in
Section III.B.2 of the preamble to todays final rule.
Overall, this analysis is consistent with EPAs conclusion that todays final rule will not have a
significant economic impact on a substantial number of marginal U.S. operators. Please see
EPAs response to EPA-HQ-OAR-2009-0923-0053-1 for further discussion about Subpart Ws
effect on marginal wells. In short, EPA has thoroughly analyzed the rules potential impact on
small entities and determined that the impact on small businesses will not be significant.
W17. (Preamble p. 118) We continue to be interested in the potential impacts of the proposed
rule on small entities and welcome comments on issues related to such impacts.
To minimize a competitive disadvantage targeting larger entities, the API does not object to
implementing the rule on small entities.
127
5 U.S.C. Sections 603(a) & 605(b).
128
5 U.S.C. Section 601(6).
129
75 Fed. Reg. 18,631.

1087
Response: EPA appreciates the commenters interest in the rules potential impacts on small
entities. The Economic Impact Analysis (Sections 5.2 and 6.3) document EPAs analysis of such
impacts.
1088
10.6
BENEFITS FOR SOCIETY

Organization: WBI HOLDINGS
Commenter:
What are the benefits of the proposed rule for society?
Comment: WBIH takes exception to EPA's statement: "A mandatory reporting system for
petroleum and natural gas systems will benefit the public by increased transparency of facility
emissions data."
Air quality permitting and compliance programs are in place, and specific facility emissions
information is already available for public review
Response: EPA disagrees with this comment. GHG emissions were not required to be measured
or reported under any federal Clean Air Act regulatory program until promulgation of 40 CFR
part 98. Although the GHG Reporting Program (40 CFR part 98) is unique, EPA considered
other federal and state programs during development of both the GHG Reporting Rule (74 FR
56260) and Subpart W, and how these existing programs treat the petroleum and natural gas
industry. EPA concluded that the addition of Subpart W to 40 CFR part 98 will supplement
rather than duplicate other U.S. government GHG programs. Preamble Section I.B discusses
EPAs authority to collect these data and the potential uses for the data, including informing both
improvements in sector based non-regulatory strategies and technologies for preventing or
reducing air pollutants, and potential policy and regulatory actions to address greenhouse gas
emissions. See Section 2 of the Economic Impact Analysis for a detailed discussion about EPAs
consideration of other federal and state GHG reporting programs.
Section III.E of the preamble summarizes the anticipated benefits of todays final rule, which
include providing the government with sound data on which to base future policies and providing
industry and the public independently verified information documenting firms environmental
performance.
Commenter:
It is necessary and appropriate to consider the underlying objectives of the requirement to
measure and report greenhouse gas emissions in the development of the underlying reporting
requirements. MidAmerican is concerned about EPAs suggestion that societal benefits of the
proposed rule include allowing citizens, community groups and labor unions to negotiate
directly with polluters to lower emissions, circumventing greater government regulation and
that publicly available emissions data also will allow individuals to alter their consumption. As
1089
an initial matter, emission reductions are an issue that should be subject to transparent debate and
should reflect the informed policy decisions of elected lawmakers and reporting should serve
these policy objectives, not the objectives of citizens, community groups and labor unions
outside the recognized legislative and regulatory processes.
Response: This comment provides an incomplete quote from the rule and therefore
mischaracterizes EPAs discussion about the rules anticipated benefits. As discussed in Section
VII.E of The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) and in
Section III.E of the preamble to todays final rule for Subpart W, EPA considered the benefits of
reporting. These Sections present the results of EPAs systematic literature review of existing
studies about the benefits of a reporting system. One example of the benefits identified by the
literature review is that a mandatory reporting system will benefit the public by increased
transparency of facility emissions data. Transparent, public data on emissions allows for
accountability of polluters to the public stakeholders who bear the cost of the pollution. Citizens,
community groups, and labor unions have made use of data from Pollutant Release and Transfer
Registers to negotiate directly with polluters to lower emissions, circumventing greater
government regulation. Publicly available emissions data also will allow individuals to alter their
consumption habits based on the GHG emissions of producers.
Please refer to Section III.E of todays final rule for a complete discussion about the anticipated
benefits for industry, government, and the public; EPAs response to EPA-HQ-OAR-0923-1074
18 for further details about the benefits.
10.7
GENERAL COMMENTS ON COSTS AND ECONOMIC IMPACTS

Organization:
Commenter: A. Potts
The problems that I see with this proposed rule are that it does nothing to tackle what is truly at
issue, imposes additional costs on business that will be passed along to the consumer, relies on
an assumption that these GHG emissions are behind the recent climate change and lack of
cooperation with business.
Response: EPA does not agree with the comment. Please see EPA-HQ-OAR-2009-0923-1303
2, for EPAs response to the comment regarding compliance costs and pass through of these
costs to consumers. Regarding the comment about the science of climate change, this rule is not
the appropriate forum for that discussion. EPA finalized findings that GHG emissions from new
motor vehicles and engines contribute to air pollution which may reasonably be anticipated to
endanger public health and welfare (74 FR 66496; December 15, 2009, Endangerment and
Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act)
(see Docket EPA-HQ-OAR-2009-0171). Prior to finalizing the endangerment finding, EPA
received over 380,000 public comments, covering the issues raised by the commenters on this
1090
reporting rule and many others. EPA published responses to these comments in the Response to
Comments documents, available at http://www.epa.gov/climatechange/endangerment.html.
Finally, EPA disagrees that the data collected under this rule would not be useful. Data
collection serves as the first step in developing sound policy. For a complete discussion about
the benefits of the rule and expected to use of the data, see EPAs response to EPA-HQ-OAR
2009-0923-1074-18.
Organization:
Commenter: A. Potts
The EPA should work in conjunction with business to come up with viable solutions for
inexpensive and effective ways to monitor the CO2 output. Over-regulation and the cost of
compliance has driven many businesses overseas seeking less stringent regulations causing
Americans lost jobs while the earth is becoming more and more polluted.
Response: This final action does not regulate GHG emissions; rather it gathers information to
inform EPAs evaluation of various CAA provisions. See Section I.C of this preamble for
complete discussion about the legal authority for the rule.
In addition, EPA has conducted extensive outreach with industry and other stakeholders to
develop reliable and cost-effective methods to monitor GHG emissions. EPA evaluated the
requirements of existing GHG reporting programs, obtained input from stakeholders, analyzed
reporting options, and developed the general reporting requirements and specific requirements
for each of the GHG emitting processes listed in Subpart W. EPA also considered public
comments it received on the April 2009 (74 FR 16448; April 4, 2009) and April 2010 (75 FR
18608; April 12, 2010) proposed rulemakings as it determined the reporting requirements issued
in todays final rule. See Section 2 of the Economic Impact Analysis for a detailed summary of
EPAs proactive communications outreach program and solicitation of stakeholder input (EPA
HQ-OAR-2009-0923). Finally, see EPA-HQ-OAR-2009-0923-1010-2 for EPAs response to
comments about its efforts to reduce the burden and costs of the rule, while still ensuring that the
program yields high quality data and essential information.
Natural Gas Distribution and Storage Are a Vital Part of the Solution for Reducing GHG
Emissions; Reporting Burdens Should be Assessed in that Context
Natural gas is the most efficient of the fossil fuels, especially when used directly by residential
and commercial consumers. Approximately 90% of the energy value of natural gas is delivered
to consumers. In contrast less than 30% of the primary energy involved in producing electricity
reaches the consumer.
1091
Natural gas is clean, efficient, abundant, and domestic. When burned, natural gas is the most
environmentally-friendly fuel because it produces low levels of unwanted byproducts (SOx,
particulate matter, and NOx) and less carbon dioxide (CO2) than other fuels. Upon combustion
natural gas produces 43% less CO2 than coal and 28% less than fuel oil.
Natural gas is also an abundant fuel. Recent prodigious discoveries of shale gas have
significantly added to this abundant resource base. Changes in economics and technology will
continue to increase our resource base estimates in the future, as they have consistently done in
the past.
Natural Gas is a domestic resource. Almost all of the natural gas that is consumed in America is
produced in North America, either in the United States or Canada, with the vast majority of that
being produced in the United States. Only a small portion1 to 2% is imported from abroad
as liquefied natural gas.
Natural gas is an essential fuel for America. The natural gas delivered by AGA members to
residential and commercial customers is consumed almost entirely to meet essential human
needsspace heating, water heating, and cooking.
Americas natural gas utilities and residential customers have led the nation in reducing the
emission of greenhouse gases over the last 40 years and can continue, with appropriate policies,
to reduce those emissions. It takes less natural gas to serve 65 million homes today than it took to
serve about half that number in 1970. Natural gas is more than just a bridge to a low carbon
future; rather, it is part of the climate change solution now and in the future, because it uses
existing technology and offers an immediate carbon reduction benefit since it has the smallest
carbon footprint of all fossil fuels.
It is important in crafting the reporting rules in Subpart W not to lose sight of this important
context for energy and environmental policy. In June 2009, AGA filed detailed comments on
EPAs April 2009 Proposed Rule for Mandatory Reporting of Greenhouse Gases (MRR), 130
including comments on proposed Subparts, A, C, NN and W. As we said then, when crafting
GHG reporting rules, it is important that EPA not inadvertently impose barriers that could keep
society from reaping the full benefit of using clean, efficient, abundant and domestic natural gas
to reduce our nations carbon footprint. Greenhouse gas reporting rules should not create
disincentives to lowering US GHG emissions by imposing unnecessary costs on the storage and
distribution of natural gas to customers, thereby raising gas utility bills and discouraging the use
of natural gas. Instead, sound public policy should encourage the efficient, direct use of natural
gas by customers to reduce overall greenhouse gas emissions.
Response: EPA disagrees that todays final rule imposes unnecessary costs on the storage and
distribution of natural gas to customers. EPA documents its assumptions and methods to
130
EPA Proposed Rule, Mandatory Reporting of Greenhouse Gases 40 CFR Parts 86, 87, 89, et al. 74 Fed. Reg.
16448 (April 10, 2009).
1092
demonstrate industry burden in the Economic Impact Analysis of the rule making docket (EPA
HQ-OAR-2009-0923) and clarifies several requirements related to natural gas distribution that
reduce the burden to much lower than many commenters assumed; please see Sections II.F and
III.B of the Preamble; EPA-HQ-OAR-0923-1016-11 also discusses EPAs analysis of costs for
the distribution segment. In addition, see EPA-HQ-OAR-2009-0923-1010-2 and EPA-HQ
OAR-2009-0923-1080-49 for EPAs response to comments regarding the rules costs and steps
the agency took to minimize the burden while ensuring the program will yield high quality data.
Furthermore, EPA has considered the potential impact of the rule on natural gas consumers; see
EPA-HQ-OAR-2009-0923-1010-2 and EPA-HQ-OAR-2009-0923-1303-2 for complete
discussion.
Finally, see todays preamble Section III.E for the benefits of this rule to society.

Commenter:
MidAmericans experience with reporting greenhouse gas emissions suggests that the
requirements and level of detail associated with a particular reporting scheme will dramatically
impact the time spent on measurement and report preparation, as well as the costs associated
with reporting. Requiring entities to focus on emissions calculations and reporting of extremely
small potential sources of emissions disproportionately impacts the costs of the reporting
requirement with very insignificant benefits.
MidAmericans experience also suggests that mandatory reporting results in higher costs. While
EPA has not proposed to require third-party verification of emissions data (a position supported
by MidAmerican), it is important to note that given the compliance certification requirements,
additional time and resources are likely to be necessary to satisfy corporate reporters and their
stakeholders that the reported data is accurate. This added burden arises not only from the direct
activities associated with reporting, but also from the indirect activities associated with
maintaining compliance management systems and auditing compliance with the reporting
program.
As noted by EPA, the initial cost of reporting (which MidAmerican believes is severely
underestimated) is significant--$56.0m in the first year. This initial $56.0m cost and subsequent
annual cost, estimated by EPA to be $21.0m, will do nothing to reduce greenhouse gas emissions
particularly when, as admitted by EPA, the benefits are very difficult to quantify and monetize.
MidAmerican supports transparent and consistent reporting; however, the level of detail should
not be so burdensome that it dramatically increases costs or risks to the reporting entity that fail
to take into account the associated benefit.
Response: EPA disagrees that the Agency significantly underestimated the costs of reporting.
See Section III.B.2 of the preamble for EPAs response to this comment and others that said the
Agency underestimated the costs; see EPAs response to EPA-HQ-OAR-2009-0923-1074-18 for
1093
a complete discussion about the benefits of the rule and expected to use of the data. Also,
Section 5 of the Economic Impact Analysis (EPA-HQ-OAR-2009-0923) presents the updated
national compliance cost estimates for todays final rule and a summary of the benefits. In
addition, see EPA-HQ-OAR-2009-0923-1010-2 for EPA response to comments about the need
to consider the burden and costs of the rule as well as the benefits.
Regarding the comment on the time and cost for a facility to comply with the rule, EPA has in
substantially reduced burden of todays final rule by simplifying the requirements for reporting
of emissions. Please see preamble Section II, for further details on the changes in the rule that
will result in significantly lower burden to reporters.
EPA appreciates the comments on third-party verification and note that todays final rule relies
on EPA verification; see Section II.D of the preamble. See also preamble Section II.N to todays
final rule promulgating 40 CFR part 98 for details about the basis for EPA verification (74 FR
56260; October 30, 2009).
Finally, EPA does not agree with the comment about GHG reductions and benefits to society. In
particular, the comment misrepresents EPAs statement that the rule benefits are very difficult
to quantify and monetize. EPAs statement refers to the challenges associated with quantifying
and monetizing the benefits, but these challenges do not mean that there are no benefits expected
from the rule. Data collection serves as the first step in developing sound policy. For a complete
discussion about the benefits of the rule and expected to use of the data, see EPAs response to
EPA-HQ-OAR-2009-0923-1074-18.

Devon Energy Corporation (Devon) has serious concerns about the U.S. Environmental
Protection Agencys (EPA) proposed rule entitled Mandatory Reporting of Greenhouse Gases:
Petroleum and Natural Gas Systems, published at 75 Fed. Reg. 18608 (April 12, 2010) and
therefore appreciates the opportunity to provide these comments. Our concerns are focused on
the unnecessarily high cost to comply with this complex rule.
Response: EPA does not agree with the comment. Please see EPAs response to EPA-HQ
OAR-2009-0923-1174-4 for a response to this comment.
Comment Excerpt Text: Devon has elected to submit these separate comments because we are
convinced of the importance of emphasizing our concern that this proposed rule will impose
1094
significant and unnecessary cost on our industry that will ultimately have negative natural gas
supply impacts at a time when US climate policy demands this low carbon fuel.
Response: EPA does not agree with the commenter. The commenter has not provided any
information regarding why or how it believes that gas supply will be impacted. Please see the
memo Subpart W Greater Economic Impact under docket EPA-HQ-OAR-2009-0923 and
EPAs response to EPA-HQ-OAR-2009-0923-1303-2 and EPA-HQ-OAR-2009-0923-1010-2 for
further discussion.
TxOGA concurs with the detailed comments submitted by the American Petroleum Institute
(API) and agrees that the U.S. EPA has significantly underestimated the burden and cost of the
proposed new mandatory reporting requirements of greenhouse gases for petroleum and natural
gas systems.
Response: EPA does not agree with the comment. Please see EPAs response to EPA-HQ
OAR-2009-0923-1151-107.
When re-structuring the rule for re-proposal, EPA should take all opportunities to simplify the
requirements and reduce burden.
Response: EPA has simplified reporting requirements in todays final rule to reduce burden.
Please see preamble Section II, for more details.
1095
VOLUME 11: DESIGNATED REPRESENTATIVE AND DATA

COLLECTION, REPORTING, MANAGEMENT, AND DISSEMINATION
11.0
DESIGNATED REPRESENTATIVE AND DATA COLLECTION, REPORTING,

MANAGEMENT, AND DISSEMINATION
11.1
DESIGNATED REPRESENTATIVE (AUTHORIZATION AND

RESPONSIBILITIES)

Reporting Contractor Emissions
i. EPA Lacks the Authority to Require Subpart W Facilities to Report Contractor Emissions
On its face, CAA Section 114 does not provide EPA with the authority to require owners or
operators of facilities, including the facilities covered by proposed Subpart W, to report GHG
emissions generated by contractors. CAA Section 114 plainly states that the Administrator may
only require monitoring, reporting, and recordkeeping submissions from any person who may
have the information necessary for the purposes set forth in [CAA Section 114(a)(1)]. 42 U.S.C.
Section 7414(a)(1). The intent of Congress is clearCAA Section 114(a)(1) requirements only
extend to those who may actually possess such information necessary for the purposes listed in
that provision. Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837,
842-43 (If the intent of Congress is clear, that is the end of the matter; for the court, as well as
the agency, must give effect to the unambiguously expressed intent of Congress.). As a general
principle, owners and operators of facilities often do not have access to the records of
independent contractors. In particular, reporting entities will often be unable to obtain from
independent contractors the detailed and often confidential information required to be reported
under proposed Subpart W, including emissions and equipment data. Therefore, EPA lacks the
authority to broadly mandate that all facilities produce and certify emissions data from
contractors.
ii. Requiring Reporting of Contractor Data is Unreasonable and Impractical
As a practical matter, EPAs concept of a designated representative and the requirement that
the designated representative certify contractor GHG emission reports is unreasonable and
impractical. The designated representative may have limited or no knowledge of the contractors
operations, the contractors potential GHG sources and associated data collected for reporting,
the quality and completeness of the data collected, quality and completeness of the verification,
and/or the environmental matters of the contractors. The proposed rule sets an unrealistic
1096
expectation for the role of the designated representative. 40 CFR Section 98.4(e)(1) includes a
certification statement containing the following language:
...I certify under penalty of law that I have personally examined, and am familiar with, the
statements and information submitted in this document and all its attachments. Based on my
inquiry of those individuals with primary responsibility for obtaining the information, I certify
that the statements and information are to the best of my knowledge and belief true, accurate, and
complete
This language sets an inappropriate standard for the owner or operator of a facility, or his/her
designated representative in terms of managing contractor data. No high-level management
official has the authority, time, or expertise to personally examine emissions information from
contractors, which is necessary to prepare the emissions report. This is particularly the case with
regards to onshore and natural gas production facilities, which may contain thousands of sources,
operated often by contractors, scattered across an entire geologic basin.
Response: With respect to comments regarding EPAs authority to require reporting of
contractor emissions, please see the response to EPA-HQ-OAR-2009-0923-1151-10.
With respect to comments about the designated representative certifying contractor emissions,
Commenter:
General Comments Certification by a Designated Representative
As proposed, this rule will require owners (i.e. Yates) of a site to report GHG emissions resulting
from rental and portable equipment located at a well head. Often, the owner (Yates) of the site
does not have emission information, or relies on information provided by a rental company (i.e.,
Compressor Systems, Inc.) that may or may not be accurate or acceptable by EPA standards.
98.4(e) requires that emissions reports be certified by a Designated Representative. Yates
Designated Representative must therefore ultimately certify emissions for units they do not
control, and there is no guarantee those emissions are auditable, verifiable, etc. Yates requests
that the EPA clarify that emissions from equipment that is on-site, but not under common
ownership, does not need to be reported by the site owner.
Response: The designated representative (DR) is the entity that is responsible for submitting the
emissions data pursuant to todays final Rule. Please see the response to EPA-HQ-OAR-2009
0923-1024-16.
11.2
PROCESS FOR DATA COLLECTION/REPORTING, MANAGEMENT, AND

DISSEMINATION
1097

Reporting, Recordkeeping and Missing Data Provisions include Requirements that Are
Infeasible, Costly, or Not Warranted to Support Mandatory Reporting Rule Objectives
Reporting and recordkeeping requirements are identified in 98.236 and 98.237. Missing data
procedures are identified in 98.235. INGAA recommends revisions or clarification to these
sections to eliminate requirements that are not practical or do not add substantive value while
incurring.
Response: EPA does not agree that the reporting and recordkeeping requirements, nor the
missing data provisions, are infeasible or impractical. See response to comment EPA-HQ-OAR
2009-0923-1151-5 and the preamble of todays final rule, Section II.F for more information.
Annual Report Due Date: Given the vast amount of information that must be collected and
compiled, IPAMS does not believe that three months is sufficient time to collect, calculate and
QA/QC the data needed to develop and submit a report. In addition, many industries are already
obligated to submit several data-intensive reports to various agencies, including EPA, in the first
quarter of the year. These include Title V semi-annual monitoring reports and annual
certifications under the CAA; quarterly deviation reports under the CAA; Discharge Monitoring
Reports under the Clean Water Act; and Tier II reports under the Emergency Planning and
Community Right-to-Know Act. IPAMS requests that EPA require annual reports to be
submitted by June 30th of each year. This date would be consistent with other registry programs,
such as that established by The Climate Registry.
Response: EPA will allow the application for the use of best available monitoring methods for
certain sources. Please see the preamble Section II.F. EPA expects reporters to meet the annual
reporting due date, which is consistent with the reporting dates in The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98); , preamble Section II.J. In addition, EPA has
modified certain methodologies for specific emission sources that will reduce burden. Please see
the preamble Section II.F.
1098
Preamble Section II.(A.) on page 18612.(KEY ISSUE) El Paso requests the EPA establish a
June 30 submittal deadline for all emission reports related to Subpart W.
Response: EPA disagrees with a June 30 reporting deadline, please see the response to
comment EPA-HQ-OAR-2009-0923-1298-76. EPA will allow the application for the use of best
available monitoring methods for certain sources. Please see the preamble Section II.F.
Organization: Southern Union Company
Commenter: Charles Wait
Of additional concern with the proposed schedule is the data warehousing of fugitive and vented
emissions. SU has a team focused on developing a GHG database; however, the focus is
primarily on combustion emissions through 2010. SU may be hard pressed to complete an
inventory of vented and fugitive sources and install these sources in a database by year's end.
EPA must realize that all source categories listed in the proposed Subpart W have no current
regulatory requirement to be considered in an emission inventory. Adding these sources to the
database will require a resource intensive ground up effort that may produce a less than accurate
inventory if full implementation is required in the first year.
Response: In certain cases, EPA is allowing for additional time to follow all of the methods in
the rule, please see Section II.F of the preamble for information on use of best available
monitoring methods. In addition, EPA has taken a number of other steps to reduce the reporting
burden (while sustaining the necessary quality of data). Please see the Section II.F of the
preamble. With the revision of the rule to allow the use of best available monitoring methods
and other reduced reporting requirements, and the use of standard spreadsheets and databases
that are readily available at retail stores, EPA does not anticipate that reporters will bear
excessive costs to complete an inventory of GHG emissions in time to record and report this
inventory by March 31, 2012, please see the Section III.B of the preamble.
Section 98.236(c) Data reporting requirements for (17) compressor wet seal degassing vents and
(18) reciprocating compressor rod packing. The total throughput of the compressor whose
emissions are reported is required under this section. This throughput is not part of the emission
estimation methods under Section 98.233 (o) and (p). In addition, Section 98.236 makes no
reference to flow meters under Section 98.234(b). As a result, API interprets the requirements
under Section 98.236(c) to allow the use of engineering estimates for determining these
throughputs. API requests confirmation from EPA that engineering estimates may be used under
Section 98.236(c).
1099
Response: Regarding the comment on compressor throughput meters, please see EPA-HQ
OAR-2009-0923-1206-63.
Commenter:
Comment on 98.236:
WBIH requests clarification on any reference to component "counts." When referencing
component counts, WBIH requests the addition of the word "leaking", i.e. 98.236(c)(19)(i)
should read, "component count for each leaking fugitive emission source."
Response: EPA has clarified the data reporting requirements for leak detection and emission
factors, to require the total number of this type of emission source found to be leaking.
Section 98.233(e) and Section 98.236(c)

Glycol Dehydrators: BP supports EPAs approach of restricting calculation of emissions to those
dehydrators without thermal control or vapor recovery on the overhead vent stream. However,
Section 98.233 (e)(2) appears to directly conflict with this approach by specifying how to
calculate emissions if the dehydrator overhead is routed to a flare or burner (thermal control). In
addition, the reporting requirements for glycol dehydrators under Section 98.236(c)(5) do not
explicitly contain an equivalent limitation or limit reporting to only those dehydrators for which
emissions must be calculated. BP requests that EPA address the conflict between Section
98.233(e) and Section 98.233(e)(2) along with making the appropriate limitation on dehydrators
would be required to report explicit in 98.236(c)(5) .
Response: EPA has reviewed the comment, and in todays final rule has clarified the Section
outlining the calculation of GHG emissions from dehydrator vents and includes calculation of
emissions from dehydrator vents without vapor recovery or thermal control devices.
General Comments Data Reporting Requirements
98.236(f) requires operators to report emissions separately for portable equipment such as
drilling rigs, dehydrators, compressors, electrical generators, etc. However, as this rule (at this
time) is only for the reporting of GHG emissions, the types of equipment emitting the GHGs is
irrelevant. Furthermore, this requirement would require installation of fuel meters on equipment
1100
that otherwise would be included in the site-wide fuel usage. PAW proposes that all fuel use at a
site be provided to the EPA as a single figure and not separate out portable equipment.
98.237(f) requires calibration reports for detection and measurement instruments used. Would
calibration reports for detection and measurement instruments also require companies to retain
records for every single pressure gauge? That is a tremendous amount of information that does
not impact GHG emissions from a site.
Response: EPA disagrees, as the primary intent of the Mandatory Reporting Rule is to inform
future policy, and EPA deems it appropriate to require separate reporting for emissions from
portable equipment. It is estimated that portable non self-propelled equipment is responsible for
over 45 percent of total emissions from onshore petroleum and natural gas production, please see
Portable Combustion Emissions docket EPA-HQ-OAR-2009-0923. In addition, the
commenter made assumptions which were not consistent with the proposed rules intent. One
was the assumption that portable equipment would need a meter to determine the volume of fuel
combusted. Reporters are not required to install a fuel meter. EPA has clarified this in todays
final rule and allows reporters to use engineering calculations based on best available data to
determine the amount of fuel combusted. EPA will also allow for certain sources best available
monitoring methods. Please see the preamble Section II.F. Finally, EPA has conducted a review
of the emissions contribution relative to reporting burden and has modified todays final rule for
onshore production and natural gas distribution, and does not include emissions from external
combustion equipment that have a rated heat capacity equal to or below 5 mmBtu/hr, please see
Equipment Threshold for Small Combustion Units docket EPA-HQ-OAR-2009-0923.
However, equipment that are at or fall below the specified mmBtu/hr level will be required to
report activity data by type of combustion equipment. With regard to retention of records for
every single pressure gauge, please see The Final Mandatory GHG Reporting Rule (Final
MRR), (40 CFR part 98); October 2009 response to comment EPA-HQ-OAR-2008-0508
0504.1 excerpt 20 in the Content of the Annual Report Section.
Commenter:
General Comments Data Reporting Requirements
98.236(f) requires operators to report emissions separately for portable equipment such as
drilling rigs, dehydrators, compressors, electrical generators, etc. However, as this rule (at this
time) is only for the reporting of GHG emissions, the types of equipment emitting the GHGs is
irrelevant. Furthermore, this requirement would require installation of fuel meters on equipment
that otherwise would be included in the site-wide fuel usage. Yates proposes that all fuel use at a
site be provided to the EPA as a single figure and not separate out portable equipment.
98.237(f) requires calibration reports for detection and measurement instruments used. Would
calibration reports for detection and measurement instruments also require YPC to retain records
for every single pressure gauge? That is a tremendous amount of information that does not
1101
impact GHG emissions from a site.

Response: EPA disagrees that all fuel use at a site be reported to the Agency as a single figure
and not separate out portable equipment. Please see the response to comment EPA-HQ-OAR
2009-0923-1015-35.
98.237: In addition to the information required by 98.3(g), you must retain the following
records:
(a) Dates on which measurements were conducted.
(b) Results of all emissions detected and measurements.
(c) Calibration reports for detection and measurement instruments used.
(d) Inputs and outputs of calculations or emissions computer model runs used for engineering
estimation of emissions.
OOC Comment: The rule should clarify what, if any, additional information must be retained
under 98.237 for offshore platforms. In the preamble, the term measurements seems to refer
to direct emissions measurements. However, the GOADS process is centered around reporting
activity data to MMS for emissions calculations and not direct emissions measurements.
Response: With regard to recordkeeping requirements, offshore reporters must retain the
information required under 98.3(g). Offshore petroleum and natural gas production must also
retain the records as set forth by MMS in compliance with 30 CFR 250.302 through 304. Not all
methods in todays final rule require data collection such as measurement, emissions detection,
or instrument calibration. Therefore reporters do not need to retain such records where not
required by methods. However reporters in state and non-Gulf of Mexico waters must calculate
their emissions using GOADS methodologies, and therefore must also retain the inputs and
outputs of those calculations. In addition, reporters subject to GOADS and reporters in state and
non-Gulf of Mexico waters must also retain the inputs and outputs of calculations required to
adjust emissions in years between GOADS cycles.
Commenter: Kim Dang
SECTION 98.237 Records that must be retained.
In addition to the information required by SECTION 98.3(g), you must retain the following
records:
(a) Dates on which measurements were conducted.
1102
(b) Results of all emissions detected and measurements. Video records of optical imaging
surveys are not required to be retained.
(c) Calibration reports for detection and measurement instruments used.
(d) Inputs and outputs of calculations or emissions computer model runs used for engineering
estimation of emissions.
Response: Regarding the suggestion on video records, EPA has clarified the rule, such that
video records of leak detection surveys with optical gas imaging instruments are not required to
be made or retained. Please see the data reporting requirements in todays final rule in Section
98.236.
Commenter: Kim Dang
Retention of Video Records. The preamble to the rule suggests that video records of optical leak
surveys showing emissions must be retained; 131 however, this requirement is not expressed in the
text of the rule itself. Moreover, it is not clear whether EPA intends for reporting entities to keep
records of all optical gas images taken at each facility, or only those images that reveal leaking
components. Kinder Morgan believes that this provision should be omitted from todays final
rule, given the space requirements for the video data and the fact that many optical imaging
devices lack a recording feature. Kinder Morgans experience is that video images from an
optical gas imaging instrument require on the order of over 3 megabytes of space per minute of
video. Video records of entire optical leak surveys will be in the gigabyte size range. Keeping
complete survey video records for multiple years at multiple facilities will require exceptionally
large electronic storage space. If EPA rejects this recommendation and elects to require retention
of optical gas imaging records, EPA should amend the text of the rule to clarify that only images
that reveal leaks need to be kept.
Response: EPA has clarified the rule, such that video records of leak detection surveys with
optical gas imaging instruments are not required to be made or retained.
98.236(c)(7)-(8): Consistent with the preceding comment, there is no need to track well
completions and workovers based on whether a well is "conventional" or '"unconventional." We
131

1103
request that EPA remove the sub-bullets (i) and (ii) from both paragraphs.
Response: EPA disagrees, and has clarified todays final rule to continue to include data
reporting requirements for well completions and workovers as identified as with and without
hydraulic fracturing. The emissions profile from these two sources is very different, and this
data is needed to inform future policy.
Organization: Canadian Association of Petroleum Producers
Commenter: Rick Hyndman
98.233 (b): CAPP noticed that the EPA has requirements for low bleed device venting referenced
in two sections: 98.233(r) and 98.233(b). Based on the redundancy of this requirement and to
guard against double reporting of emissions CAPP recommends that the requirements only be
reported in one section.
Response: EPA has clarified todays final rule such that natural gas driven pneumatic devices
have been consolidated to all report under Section 98.233(a).
CAPP also proposes the following rewording of 98.233(n)(8): "Any emissions calculated under
this source are excluded from other emission sources in 98.233" and that if a source is covered
elsewhere but is sent to flare, reporters be given the choice of where / how to report it, based on
ease of reporting.
Response: EPA has reviewed the comment and has clarified Section 98.233(n) of todays final
rule such that flare emissions determined under paragraph (n) must be corrected for flare
emissions calculated and reported under other paragraphs to avoid double counting of these
emissions.
Organization: Contek Solutions, LLC
Commenter: Jim Johnstone
Reporting: Will the EPA be sending out standardized report forms for operators to fill out
Response: No, EPA will not be issuing reporting forms. The reports must be submitted
electronically. Please see The Mandatory Reporting of Greenhouse Gases Rule, (Final MRR),
(40 CFR part 98) preamble Section II.A.
1104

Aka recommends deleting 98.236(d). This paragraph (d) requests minimum, maximum and
average throughput for each operation listed in paragraphs (a)(1) through (a)(8) of this section.
This is a very vague request and has no apparent meaning for most of the operations described in
(a)(1) through (a)(8). In addition, this information does not appear to be useful in determining or
understanding the GHG emissions information that is otherwise reported.
Response: EPA has revised 98.236(d) in todays final rule to only require reporting of annual
throughput, as determined by engineering estimates based on best available data.
11.2.1 DATA COLLECTION METHODS (COMMENTS ON SECTION VI.B OF THE

PREAMBLE)
Organization: Western Resource Advocates
Commenter: Robert Harris
Source data on federal and state public lands should be aggregated by the relevant management
area, or disaggregated, so as to be useful to land management agencies.
EPA should collect information on the GHG emissions of onshore production facilities located
on public lands in a form that may be useful to other agencies conducting relevant analyses,
including under the National Environmental Policy Act (NEPA), 42 U.S.C. 4321 to 4370f.
The American Association of Petroleum Geologists definition of a basin may not adequately
assist land management agencies in meeting their statutory obligations. For example, the basin
boundaries may not match Bureau of Land Management (BLM) field office boundaries or the
boundaries of national forests. Because the Proposed Rule would require reporting only
aggregated emissions data, land management agencies may not be able to readily divide the data
for an accurate estimate of GHG emissions occurring on only part of a basin within the relevant
national forest or management area. Therefore, emissions reports may not be helpful for
conducting or evaluating resource management plans or forest plans prepared under the Federal
Land Policy and Management Act, 43 U.S.C. 1701 to 1785, or the National Forest
Management Act, 16 U.S.C. 1600 to 1614, respectively.
To remedy this problem, EPA could require aggregated reporting of GHG emissions from oil
and gas facilities on federal public lands to match the boundary of the relevant BLM field office
or national forest. Alternatively, the Proposed Rule could require operators to show their work
by attaching disaggregated well pad data as an appendix to the aggregated basin-wide report.
1105
According to EPA, the agency did not reject the well pad option because it was infeasible, but
rather because the considered reporting thresholds would result in either too many or too few
reporters. 75 Fed. Reg. at 18615.
Other agencies could use the disaggregated location-specific well pad data to help quantify GHG
emissions in the relevant BLM field office, national forest, or other geographic area. EPAs
maintenance of a publicly accessible and searchable database, capable of providing flexible data
outputs, would further maximize the utility of information collected by the EPA under the
Proposed Rule by minimizing inefficient and duplicative efforts by government agencies and
members of the public.
Response: Regarding the comment on EPA collection and sharing of data with other Federal,
State, and regional programs, please see the Mandatory Reporting of Greenhouse Gases Rule,
(Final MRR), (40 CFR part 98), preamble Section II.O. The petroleum and natural gas
systems monitoring and reporting requirements in todays final rule, are being added to the GHG
RP and will gather data to inform future policy and programs. EPAs decision to use American
Association of Petroleum Geologists definition of a basin, the EIA field definition, and data
reporting level are discussed in the Technical Support Document (TSD) for todays final rule
found in docket (EPA-HQ-OAR-2009-0923). Todays final rule requires reporting at the basin
level in order to manage burden and maintain consistency of the data received, and therefore
cannot introduce additional facility definitions such as the suggested BLM field office
boundaries, due to the additional burden and it will not increase the quality of data nor can EPA
possibly take into account all possible reporting configurations that various parties desire and
therefore basin-level reporting is the most appropriate option.
11.2.1.1
ELECTRONIC SIGNATURES
11.2.1.2
USE OF UNIQUE IDENTIFIERS
11.2.1.3
METRIC UNITS
11.2.1.4
DELEGATION OF AUTHORITY TO STATES FOR DATA COLLECTION
1106
11.2.1.5
USE OF AN ELECTRONIC REPORTING SYSTEM

Report submittal. EPA proposes that all reports will be submitted electronically. This may be
burdensome on some small businesses that don't have the staff, expertise or access to submit the
data electronically. EPA should allow small businesses to submit a hard copy of the required
information.
Response: Please see Section V.B of the preamble of The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98), promulgated on October 30, 2009, for a response to this
comment.
11.2.2 DATA QA AND FEEDBACK BY EPA TO REPORTERS
11.2.3 DATA DISSEMINATION
11.2.3.1
DATA DISSEMINATION TO THE PUBLIC

Organization:
Commenter: J. Andes
Clearly, the oil and gas industry will not voluntarily report this information to the American
public. Therefore, the industry must be obligated by federal regulation and enforcement to make
public the information
0923-1015-27.
Organization:
Commenter: C. Taylor
1107
I further urge that the industry's reported emissions are made public as soon as possible, and that
the combination of greenhouse gas emissions and pollution from accidents and spills be made
public so that all the necessary changes in the regulation of this industry can be made NOW!
0923-1015-27.
Organization:
Commenter: K. Wolney
Your data collection methods should be unimpeachable and the results should be reported to the
public so we can see the consequences of our dependence on oil.
0923-1015-27.
11.2.3.2
SHARING OF DATA WITH OTHER STATE AND FEDERAL AGENCIES

Organization:
Commenter: Anonymous
Public Citizen has noted that the Federal Energy Regulatory Commission; formerly the Natural
Gas Act; illegally ruled in March 2004 that states have limited jurisdiction over the permitting
and siting of Liquified Natural Gas facilities inside their borders. Since state and local
governments already have inadequate input into these projects, how does a (proposed) federal
regulation that will not require control of greenhouse gases; only requiring sources above
certain threshold levels, to monitor and report emissions help local and state governments?
Response: For a response to this comment, please see Sections I.E, II.O, and V of the
preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98);
promulgated on October 30, 2009.
GHG emissions data collected under the Proposed Rule should be in a format that can readily
assist agency efforts to reduce GHG emissions.
Proper GHG emission data reporting on public lands could also assist sister agencies in their
1108
efforts to reduce GHG emissions. For example, on September 14, 2009, Department of Interior
(DOI) Secretary Ken Salazar announced the creation of the DOI Carbon Footprint Project
(Project). The Project will develop a unified greenhouse gas emission reduction program,
including setting a baseline and reduction goal for the Departments greenhouse gas emissions
and energy use. U.S. Dept. of Interior, Sec. Order No. 3289.3 By tailoring the Proposed Rules
reporting requirements with respect to facilities on public lands, EPA could help DOI establish
an accurate baseline for existing GHG emissions on public lands as well as track progress toward
accomplishing DOIs emission reduction goals.
Response: For a response to this comment, please see Sections V.A and V.B of the preamble of
The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), promulgated on
October 30, 2009.
1109
VOLUME 12: MONITORING AND QA/QC REQUIREMENTS
12.0
MONITORING AND QA/QC REQUIREMENTS

Commenter: Kim Dang
As EPA has acknowledged, fugitive and vented GHG emissions from the petroleum and natural
gas sector are inherently difficult to monitor and quantify. This difficulty is a function of both the
facilities themselves, which are extremely complex and contain thousands of disperse and
difficult-to-access components, and of the elusive nature of the emissions. Kinder Morgan
described these problems in its comments on the original Subpart W proposed in spring 2009,
noting that direct measurement of fugitive and vented emissions is inaccurate and unreasonably
costly. 132
Kinder Morgan appreciates EPAs consideration of industry comments on the original Subpart
W proposal, and its attempt to address some of the problems with that original proposal in the
newly proposed rule. As a general matter, Kinder Morgan supports EPAs proposal to require
reporting only from those components of petroleum and natural gas facilities that are responsible
for 80% of GHG emissions. 133 Compared to the original Subpart W, this streamlined approach
should reduce the amount of effort and cost wasted on quantifying relatively trivial emission
sources. In addition, Kinder Morgan strongly supports EPAs proposal to make use of
engineering estimation and emission factors where possible. 134 These methods generally provide
reliable emissions, are well-understood and widely used within the industry to create GHG
emission profiles, and are far more practical and cost-effective than individual direct
measurement at millions of individual components.
Response: EPA has reviewed and considered this comment.
Commenter: Kim Dang
SECTION 98.234 Monitoring and QA/QC requirements.
132
See Comments of Kinder Morgan on the original Subpart W, Document ID No. EPA-HQ-OAR-2008-0508-0370
(filed June 9, 2009).
133
See Proposed Subpart W, 75 Fed. Reg. at 18,614.
134
See Proposed Subpart W, 75 Fed. Reg. at 18,620 (noting that direct measurement is called for only where no
credible engineering estimation methods or emissions factors exist that can accurately characterize the
emissions.).
1110
(a) You must use the method described as follows to conduct annual leak detection of fugitive
emissions from all source types listed in SECTION 98.233(p)(3)(i) and (q) in operation or on
standby mode that occur during a reporting period.
(1) Optical gas imaging instrument. Use an optical gas imaging instrument for fugitive emissions
detection in accordance with 40 CFR part 60, subpart A, SECTION 60.18 (i)(1) and (2) of the
Alternative work practice for monitoring equipment leaks. In addition, You must operate the
optical gas imaging instrument to image the source types required by this proposed reporting rule
in accordance with the instrument manufacturers operating parameters. In addition to the optical
gas imaging instrument, alternative technologies may be used for fugitive emissions detection in
accordance with a testing method published by an industry consensus standards organization
(e.g., ASTM, ASME, API, GPA, etc.) and instrument manufacturer recommendations. These
include, but are not limited to, organic vapor analyzers, portable hydrocarbon gas detectors,
ultrasonic leak meters, ultrasonic acoustic detectors, and soap solution.
(2) [Reserved]
(b) All flow meters, composition analyzers and pressure gauges that are used to provide data for
the GHG emissions calculations shall use measurement methods, maintenance practices, and
calibration methods, prior to the first reporting year and in each subsequent reporting year using
an appropriate standard method published by a consensus standards organization such as, but not
limited to, ASTM International, American National Standards Institute (ANSI), and American
Petroleum Institute (API). If a consensus based standard is not available, you must use
manufacturer instructions to calibrate the meters, analyzers, and pressure gauges.
(c) Use calibrated bags (also known as vent bags) only where the emissions are at near
atmospheric pressures such that it is safe to handle and can capture all the emissions, below the
maximum temperature specified by the vent bag manufacturer, and the entire emissions volume
can be encompassed for measurement.
Response: EPA has revised todays final rule to allow specific alternative leak detection
methods. Please see the preamble Section II.F and response to comment EPA-HQ-OAR-2009
0923-1039-18.
Section 98.234 Monitoring and QA/QC requirements
In reviewing the scope of the GHG Monitoring Plans, we are concerned about the workload
associated with developing and maintaining the GHG Monitoring Plans. The scope of the
monitoring plans make sense at the traditional CAA facility level, however the burden associated
with keeping them up to date, coupled with their diminished value as their size increases could
make them essentially useless at the hydrocarbon basin level. We would support breaking the
1111
geographic extent of the monitoring plan scope down to a lower, more operationally relevant
basis, such as the sub-basin groupings described previously.
Response: EPA disagrees with the comment regarding the monitoring plan. Please see the Final
Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.I.
12.1
METHODOLOGY USED TO DETECH GHG EMISSIONS

Organization:
Commenter: J. Sizelove
EPA should require direct measurement of emissions. Using equations to estimate emissions is
inaccurate. We need direct measurements to ensure the data is accurate to develop the best
policy.
Response: EPA disagrees that direct measurement is appropriate for all emissions sources or the
most cost-effective way to inform future policy. Each source in todays final rule has a carefully
considered methodology that is sufficiently accurate and cost-effective to provide the
information EPA needs for future policy. In some cases, engineering models or calculations are
more accurate than spot direct measurements, and continuous measurement on the enormous
number of individual emission sources is not at all practical or necessary. For further information

Alternatives to the Optical Gas Imaging Instrument Should Be Allowed for Leak Screening
As discussed in Comment XII, the current lack of trained and experienced infrared camera
operators and GHG fugitive emissions survey companies required to complete the annual leak
inspections necessitates a phased-in approach. Furthermore, limited camera technology options
exist, severely limiting the options for compliance. INGAA recommends the acceptance of
alternatives.
Leak screening and detection can be accomplished using a variety of lower cost tools and
methods. The Proposed Rule prescribes an optical camera to identify leaks during the annual
facility leak survey. INGAA recommends allowing alternatives to the optical camera for leak
detection. Standard methods are available, including methane detectors, ultrasonic/acoustical
instruments, soap solution, etc. In most applications, these methods have been used in practice
for a longer period than the optical gas imaging camera technology and are equal to or better
1112
than infrared optical imaging camera technology for locating accessible leaking components.
INGAA requests that the Final Rule include all demonstrated technologies and alternatives that
are used in practice. Limiting the technology choice to optical infrared cameras is unreasonably
restrictive, and unintentionally implies favoritism or endorsement of a single technology by
precluding proven, less expensive leak detection and measurement instruments and/or methods.
It is also important to note that the optical infrared camera only indicates the presence of a leak.
Therefore, the optical camera, Method 21, and soap solution perform the identical function of
identifying leaks. The camera does provide the advantage of surveying difficult to access (e.g.,
elevated) components, as other methods require direct access to the component.
A rule that allows alternative methods, with consensus methods validated separate from the rule
and industry practices used in the interim, provides the most expedient path to support rather
than preclude innovation and technology advancement. Limiting instrumentation or prescribing
hierarchy for monitoring and measurement introduces unnecessary restrictions which preclude
knowledgeable professionals from implementing the right technical approach for leak
monitoring.
It is INGAAs understanding that the FLIR GasFindIR camera is currently the primary optical
infrared camera technology manufactured in commercial quantities. With limited market
competition, EPA has inappropriately endorsed a technology vendor and created a demand that
will preserve or even raise the cost of this technology. Alternatives based on industry standard
practices in the near-term and peer-reviewed, consensus (e.g., ASTM) methods in the long-term
should be allowed.
In addition to substantial instrument cost, camera technology performance is susceptible to
environmental factors. The camera is not intrinsically safe within an enclosed environment (e.g.,
compressor building) and typically requires a hot work permit to conduct a leak survey. The
expertise required to operate the camera involves significant training and experience with
imaging at similar industry sources, whereas other technologies and methods such as soap
solution can be easily applied and observed by less-experienced personnel. EPA should not
select and promote a single preferred technology when suitable and viable alternatives are
available and used in practice.
EPA has funded field studies that have employed all of the recommended technologies and leak
detection methodologies discussed in these comments. These tools provide the necessary
flexibility to screen leaking components through use of accepted and practiced techniques that
may be more suitable based on environmental conditions, available tools, and/or survey team
experience level.
Similar to the October 2009 Final Rule, industry practices or consensus standard should be
permissible. For example, the 2009 Final Rule references industry practices in many subparts,
such as Subpart MM for Suppliers of Petroleum Products which indicates at 98.394(a)(1)(ii):
(ii) Where no appropriate standard method developed by a consensus-based standards
organization exists, industry standard practices shall be followed.
1113
Since methods are not consistently available for leak surveys of natural gas systems, similar
language should be added to Subpart W. EPA should allow emissions detection and
measurement methods developed and relied on in practice, including use in EPA-sponsored
projects and for emissions reported under the Natural Gas STAR program. INGAA recommends
adding language in 98.234(a)(2) [currently reserved] to include the following:
In addition to the optical gas imaging instrument, alternative technologies may be used to
conduct annual leak detection of fugitive emissions in accordance with industry standard
practices, consensus-based standards, or manufacturer recommendations. These include, but are
not limited to, organic vapor analyzers, portable hydrocarbon gas detectors, ultrasonic leak
meters, ultrasonic acoustic detectors, and soap solution.
As discussed in these comments, the optical gas imaging instrument has limitations that are best
resolved through the inclusion of alternative technologies, practices, and standards.
Optical Gas Imaging Instrument Procedures and References to the AWP Should Be Revised
98.234(a)(1) specifies the use of an optical gas imaging instrument for fugitive emissions
detection in accordance with 40 CFR part 60, Subpart A, 60.18(i)(1) and (2) of the Alternative
Work Practice (AWP) for monitoring equipment leaks. INGAA supports EPAs decision to limit
and minimize the AWP requirements listed in 60.18. However, this AWP was not intended to
address methane or GHG leaks and contains many provisions and requirements that are
inappropriate or too restrictive for natural gas transmission and storage component leak
screening.
For example, 60.18(i)(1) specifies the optical gas imaging instrument specifications. This
section precludes the use of the commercially available Remote Methane Leak Detector
(RMLDTM) because it does not provide an image of the potential leak as defined in (1)(i) or
video records as defined in (1)(ii).
60.18 (i)(2) provides a detailed procedure for daily instrument checks that was intended to
address mixed hydrocarbon streams and varied flow rates from refinery processes. This process
scenario is not applicable to natural gas transmission and storage sources. Natural gas
composition is well defined, and is not subject to the significant composition variability
considered under this section. Furthermore, the mass flow rate determination [paragraph
(i)(2)(i)(B)] using monitoring frequency and mass fraction of detectable chemicals is also
inappropriate for natural gas transmission and storage sources.
INGAA recommends that reference to 60.8(i)(2) be replaced with an alternative calibration
approach that is more appropriate for application to natural gas sector sources for methane leak
surveys. For example, flow could be modulated using a rotameter selected to provide a threepoint check over a range of flows using a 100 percent methane standard. In addition, a
calibration check could be completed using the available high volume sampler calibration
standards per 98.234(d)(4) and this section should be cited under 98.234(a)(1). The calibration
criteria under 98.234(d)(4) indicates the following:
1114
Calibrate the instrument at 2.5 percent methane with 97.5 percent air and 100 percent CH4 by
using calibrated gas samples and by following manufacturers instructions for calibration.
The Proposed Rule referenced only certain sections of the AWP and eliminated the AWP leak
practice procedure. INGAA recommends instrument operation in accordance with the
manufacturers recommendations or industry standard practices. In addition, the General
Provisions under 40 CFR Part 60 Subpart A Section 18 Paragraph (a) should be revised to
include reference to leak screening under 40 CFR Part 98.
The Rule Should Include Provisions for Eliminating or Decreasing the Frequency of Leak
Screening
The requirement for an annual survey is basically an arbitrary decision since data are not
available to support the proposed frequency required for surveys of natural gas transmission and
storage facilities. In addition, portions of the facility that are not in vibration or heat-cycle
service may be amenable to much less frequent surveys. Thus, the Proposed Rule should include
provisions for eliminating or decreasing the frequency of annual leak screening should data
and/or EF advances meet accuracy needs and leak screening data indicate that intervals can be
relaxed. Since data gathering for reporting will provide a significant influx of information on
leaks from natural gas systems, fugitive leak detection should not be adopted in perpetuity.
Annual leak surveys should be phased out over time as data are collected and assessed,
alternatives such as improved equipment- or component-based emission factors are identified,
and reasonable data quality and inventory accuracy objectives are met.
This could be implemented through text additions to the rule, or citations to procedures in the EF
reference document discussed above that could be used to identify alternative approaches (e.g.,
population-based emission factors that preclude a leak survey and identification of facility
systems (e.g., yard piping) that do not warrant annual surveys).
Response: Concerning the availability of equipment or trained operators, for certain sources
EPA will allow the application for the use of best available monitoring methods. Please see the
preamble Section II.F.
EPA has revised todays final rule to allow specific alternative leak detection methods. Please
see the preamble Section II.F.
EPA disagrees with the comment to allow reporters to choose methods for leak detection that are
not specified in subpart W. Please see The Final Mandatory GHG Reporting Rule (Final
MRR), (40 CFR part 98) preamble Section L. EPAs analysis of leak detection methods is
outlined in the April 2010 Technical Support Document (TSD) found in docket (EPA-HQ-OAR
2009-0923-0027), and in the April 2010 preamble Section II.E.
EPA disagrees that it has inappropriately endorsed a technology vendor. Todays final rule
allows for the use of any vendor of optical gas imaging technologies that meet leak detection
specifications of the Alternative Work Practice to Method 21, any other Method 21 leak
1115
detection technologies, infrared laser beam illuminated instruments, or acoustic leak detection
instruments.
EPA recognizes the importance of ensuring safety, and has added alternative methodologies to
ensure safety. Please see response to comment EPA-HQ-OAR-2009-0923-1024-11.
EPA disagrees with the commenter that instrument specifications in the Alternative Work
Practice to Method 21 for 40 CFR part 60 are inappropriate for the natural gas sector. EPA 40
CFR part 60 regulations are also applicable to the natural gas processing sector, and todays final
rule will continue to use these standards.
EPA disagrees with eliminating or decreasing the minimum annual leak detection survey to
calculate emissions. Annual reports must be calculated and submitted each year. Please see The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble, Section II.H.
EPA disagrees with the comment on how to update emission factors. Please see the response to
EPA-HQ-OAR-2009-0923-1299-5.
Concerning manufacturers recommendations, please see the response to EPA-HQ-OAR-2009
0923-1042-27.
Organization: Delmarva Power a PHI Company
Commenter: Wesley L. McNealy
Leak Survey Requirements
In addition to DPL's concerns that Subpart W's proposed leak survey requirements would be
duplicative, EPA's proposal prompts other concerns. Specifically, Subpart W requires the use of
optical scanning through the use of infrared cameras for leak detection purposes. Such equipment
is costly ranging from $80,000.00 - $100,000.00 per camera. In light of the fact that most LDCs
would require more than one camera for EPA's proposed leak detection requirements, LDCs
would not only be forced to incur the cost of multiple cameras but would also incur Significant
costs associated with employee training and maintenance. In contrast, PHMSA allows LDCs to
utilize a full range of leak detection equipment and materials for each circumstance such as gas
detection wands and leak detection solutions; DPL recommends that the EPA consider this
option.
Response: EPA disagrees that the rule is duplicative of DOT regulations. Please see response
to comment EPA-HQ-OAR-2009-0923-0955-7. EPA disagrees with the comment on cost.
Please see response to comment EPA-HQ-OAR-2009-0923-0049-7 and the preamble Section
III.B. EPA has revised todays final rule to allow specific alternative leak detection methods.
Please see the preamble Section II.F and response to comment EPA-HQ-OAR-2009-0923-1039
18.
1116

In light of EPA partners recommendations under the Natural Gas STAR Program, including for
the use of ultrasound to detect valve leaks at production facilities, 135 EPAs pessimistic
characterization of measures to detect and fix leaks at production facilities in this rulemaking, 75
Fed. Reg. 18608, 18623, appears contrary to the facts before the agency.
Estimating fugitive emissions of production facilities solely through population count and
emission factors is unlikely to meaningfully aid operators in reducing GHG emissions
and conserving hydrocarbons. Proposed section 98.233(r) would generically calculate
fugitive emissions from production facilities using EPAs default estimates provided in
proposed Table W-1. Entirely divorced from site-specific conditions, this calculation amounts to
little more than a paperwork exercise. Instead, EPA should mandate annual leak detection at
relevant onshore production source types listed in section 98.232(c) as a common-sense measure
to help operators improve efficiency and reduce their GHG footprint.
Response: EPA disagrees with the comment to require leak detection for onshore production, as
population emission factors result in reduced burden on industry while maintaining the necessary
quality of data to inform policy. Please see the preamble Section II.F. EPAs analysis of the use
of population emission factors is outlined in the Technical Support Document (TSD) found in
docket (EPA-HQ-OAR-2009-0923-0027), and in the April 2010 preamble Section II.E.
Section 98.233.q requires a leak detection survey of connectors, block valves, control valves,
compressor blowdown valves, pressure relief valves, orifice meters, other meters, regulators, and
open ended lines using an optical gas imaging instrument at compressor stations. The leak
detection survey as proposed is time consuming and expensive. For a detail of issues on leak
surveys please see above comment #2 under General Comments on Subpart W. If leak surveys
are required, NMGC proposes that alternative methods of detecting leaks, such as the soap
bottle, be allowed. They are more effective than infrared cameras and much less costly to
employ.
0923-1039-18. EPA disagrees with the comment on cost. Please see response to comment EPA
HQ-OAR-2009-0923-1099-8 and the preamble Section III.B.
135
PRO Fact Sheet No. 602 (Natural Gas STAR partner Texaco estimated that using ultrasound to identify valve
leaks will generally pay for itself in less than a year).
1117

Allow Broader Range of Leak Detection Equipment for Leak Surveys at Underground Storage
Facilities.
Similarly, EPA should not require the use of optical gas scanning equipment (infrared FLIR
cameras) but should allow a broader range of effective leak detection equipment for leak surveys
at underground storage facilities, as we argue elsewhere in these comments. Consumers
appreciates the opportunity to comment. We look forward to working with you to make the
reporting rule for natural gas utilities practical and effective. If you should have any questions,
please call me at 517-788-2201.
0923-1039-18.
Allow Broader Range of Leak Detection Equipment--To reduce costs and to improve results, Consumers urges EPA not to require the use of optical
gas scanning equipment/ infrared cameras, and instead to allow the use of a full range of
available leak detection equipment, allowing the operator to select the tool that is most effective
for a given situation. The best way to do this may be to reference existing federal and state leak
detection regulations and relevant portions of the industry standard guidance for leak surveys.
0923-1039-18.
(Preamble p 75) EPA is seeking comments on allowing the OVA/TVA to be used as another
option to the optical imaging camera in this proposed rule.
API supports the use of OVA/TVA as an alternative option to the optical imaging camera. The
leak detection requirements should also be consistent with existing leak detection and repair
1118
(LDAR) regulations for onshore natural gas processing facilities to minimize additional burden
required to collect data on these components. In addition, other alternatives should be allowed
for leak detection. Limiting leak detection to the optical camera alone is too restrictive and will
result in increased costs, as it will immediately create a shortage of qualified personnel due to the
large number of entities required to perform this leak monitoring. Existing leak detection and
repair (LDAR) programs using OVAs should explicitly be allowed for Subpart W. Additional
accepted alternatives such as methane detectors, ultrasonic / acoustical methods, and soap
solution are available, are already used in practice, and provide better or equivalent data in many
cases. These methods should also be included in Section 98.234(a), and where standard methods
are not readily available, industry standards or best practices should be allowed to accommodate
those entities that cannot obtain the required equipment or manpower resources within the
implementation timeline of this rule.
0923-1039-18. Concerning the availability of equipment or trained operators, for certain sources
preamble Section II.F. EPA disagrees with the comment on cost. Please see response to
comment EPA-HQ-OAR-2009-0923-0049-7 and the preamble Section III.B.
Organization: The Dow Chemical Company
Commenter: Robert Rouse
Dow Supports Allowing the OVA/TVA to be Used as Another Option to the Optical Imaging
Camera in this Proposed Rule.
Dow operates a natural gas storage facility that is associated with a larger petrochemical facility.
Therefore, proposed 98.233(q) requires the use of the methods described in proposed 98.234(a)
to conduct an annual leak detection of fugitive emissions from the fugitive sources that are listed
in the proposed rule. Rule 98.234(a)(1) requires the use of an optical gas imaging instrument in
accordance with 40 CFR 60, subpart A sections 60.18(i)(1) and (2) Alternative work practice for
monitoring equipment leaks.
Dow generally supports the use of an optical gas imaging instrument on a voluntary basis to
detect leaks. However, as solicited by EPA on Page 18623 of the April 12, 2010 Federal Register
proposal, Dow comments that allowing the use of an OVA/TVA to be used as another option
should be included in the final rule for the following reasons:
1. A number of OVA/TVA instruments are readily available to our on-site fugitive emission
monitoring personnel for annual monitoring and any desired follow-up monitoring.
2. The OVA/TVA instrument is capable of measuring a concentration of a leak, if detected, and
Dow suggests that an instrument reading > 10,000 ppmv as methane should be treated as a leak
for the purpose of this reporting rule.
1119
3. The alternative work practice contains burdensome recordkeeping provisions (Reference 40

CFR 60.18(i)(4)(vi) where a video record must be used to document the leak survey results. The
video record must include a time and date stamp for each monitoring event. In addition, the
video record must show that each regulated fugitive emission source can be identified. All video
records must be kept for five years. Dow has previously expressed concerns with this burden of
video records in our comments on the proposed rule to establish the Alternative Work Practice.
In support of using an existing OVA/TVA instrument, the recordkeeping requirements are easier
to implement and can be readily added to existing recordkeeping systems.
0923-1039-18. EPA has revised the rule and videos are not required. Please see response to
comment EPA-HQ-OAR-2009-0923-1024-49.
In addition, the Alternative Work Practice associated with the use of an optical gas imaging
instrument requires that the regulated equipment also be monitored annually using a 40 CFR Part
60, Appendix A-7, Method 21 monitor at the leak definition required in an applicable subpart
(Reference 40 CFR 60.18(h)(7)). Thus, the Alternative Work Practice technically requires annual
monitoring with an OVA/TVA instrument. If EPA's intent in this proposed rule is to only require
an annual survey of impacted sources with an optical gas imaging instrument, then the final rule
should clearly reflect this intent with respect to the AWP.
0923-1039-18. EPA disagrees that subpart W annual leak detection surveys conducted with an
optical gas imaging instrument must also be surveyed annually using Method 21, as this is not
specified in subpart W.
We assume EPA means infrared cameras which are very expensive to purchase at nearly
$100,000 each, or that rent for upwards of $2,900 per week. Use of this camera on this number
of stations in our system would be burdensome. New Mexico Gas Company rented one of these
cameras last year to investigate their potential for identifying leaks when this rule was first
proposed. Our company found that the results obtained are very dependent on the skill of the
operator and how the camera is utilized plus the ambient conditions at the time of use. Our
1120
experience with this equipment leads us to believe that traditional leak detection methods yield
far better and consistent results.
Response: Please see response to comment EPA-HQ-OAR-2009-0923-1099-20.
Leak Detection Survey - Section 98.233.q requires a leak detection survey using an optical gas
imaging instrument for sources listed in 98.232.e.7 and 98.232.i.1 (fugitive emissions at
compressor stations and LDC's). The leak detection survey as proposed is time consuming and
expensive. First, optical gas imaging instruments are not necessarily more effective than other
methods of leak detection. In fact our tests show them to be vary [sic] susceptible to the operator
and ambient condition changes. The preamble states (pg 18623) that optical gas imaging
instruments are "able to scan hundreds of source types quickly". In our experience with infrared
cameras this is not true. We had to stop at each piece of equipment individually to determine if
there was a leak. For the camera to provide useful results one must be within 3ft of the source
and have little to no wind. Second, despite EPA referencing methods for performing the leak
survey in "Alternative work practice for monitoring equipment leaks", the results will vary
widely since some parameters are not specifically defined.

In addition, API requests that alternatives to the optical camera be allowed for leak detection
from reciprocating compressor rod packing. Limiting leak detection to the optical camera alone
is too restrictive and will result in increased costs, as it will immediately create a shortage of
qualified personnel due to the large number of entities required to perform this leak monitoring.
Accepted alternatives such as methane detectors, ultrasonic / acoustical methods, and soap
solution are available, are already used in practice, and provide better or equivalent data in many
cases. These methods should be included in Section 98.234(a), and where standard methods are
not readily available, industry standards or best practices should be allowed to accommodate
those entities that cannot obtain the required equipment or manpower resources within the
implementation timeline of this rule.
There should also be additional metering and monitoring technologies made available for
measuring the leaks from reciprocating compressor rod packing other than those currently
identified in Section 98.234(c) and (d). There are a variety of monitoring devices that could be
1121
used to measure flow and emissions that would still meet the accuracy requirements of the
proposed rule.
EPA has revised the rule to allow acoustic leak detection and quantification for compressor
venting (through valve leakage). EPAs analysis of direct measurement methods is outlined in
the Technical Support Document (TSD) found in docket (EPA-HQ-OAR-2009-0923-0027), and
in the April 2010 preamble Section II.E.
Commenter:
Comment on 98.234(2):
This section should provide flexibility for leak detection monitoring through the use of any
industry accepted practices, such as soap solution, instead of optical gas imaging instruments.
0923-1039-18.

Organization: Auburn Environmental
Commenter: Jim Riffle
I strongly agree with the Alternate Work Practice (AWP) outlining the usage of optical gas
imaging equipment for leak detection. I have found that optical gas imaging greatly expedites
leak detection surveys and allows hundreds of components per shift to be evaluated due to the
elimination of the need for aerial man-lifts, fall protection, and ladders to reach elevated
components. I have personally witnessed projects where large VOC emissions were completely
overlooked by TVA, OVA, and bubble solutions methods as part of Leak Detection and Repair
(LDAR) programs.
I strongly disagree with public comments that the usage of optical gas imaging cameras would be
cost prohibitive. Many contractors are now offering leak detection service including optical gas
imaging at very affordable prices that would eliminate the need for companies to purchase
cameras and train individual in the proper usage. In addition, the amount of man-hours required
to perform a leak detection survey is greatly reduced by implementing optical gas imaging as
part of an LDAR program.
I believe that the usage of the optical gas imaging camera will reduce overall operating costs
associated with leak detection, enhance safety, reduce fire risk, and promote Greenhouse Gas
(GHG) reduction.
1122
Response: EPA has reviewed and considered this comment. Regarding the comment on costs,
please see the preamble Section III.B and the Economic Impact Analysis Section 5.
Organization: The Clean Energy Group
Commenter: Michael Bradley
Measurement Methods
As noted above, we support EPAs proposal to scale back the number of sources that are
required to conduct direct emissions measurement, and agree with the following technologies
that EPA proposes to be used for direct measurement of those large sources: high volume
samplers, meters (such as rotameters, turbine meters, hot wire anemometers, and others), and/or
calibrated bags.
However, the Clean Energy Group is concerned with the cost burden associated with EPAs
proposed technology for fugitive emissions detection (e.g., optical gas imaging instrument) for
processing, transmission compression, underground storage, LNG storage, LNG import and
export terminals, distribution, and above ground M&R and gate stations. The proposed rule
requires that on an annual basis, the entire population of fugitive emissions sources proposed for
reporting in this rule would be surveyed at least once. EPA should provide more flexibility by
allowing the use of population emission factors for these facilities. Where population emission
factors cannot be used, EPA should allow the use of Organic Vapor Analyzers (OVA) or Toxic
Vapor Analyzers (TVA).
EPA states that the OVA/TVA requires the operator to physically access the emissions source
with the probe and thus is much more time intensive than using the optical gas imaging
instrument. EPA also notes that the OVA/TVA range is limited to the reach of an operator
standing on the ground or fixed platform, thus excluding all emissions out of reach.
The Clean Energy Group disagrees with the proposed exclusion of OVA/TVA from use in the
MRR. Industry has experience with and currently uses OVA/TVA to identify fugitive emissions.
The industry finds OVA/TVA reliable and accurate. Given the fact that most equipment is at
ground level, the majority of the fugitive emissions are always within reach. The Clean Energy
Group recommends that EPA allow the use of OVA/TVA as another option to the optical
imaging camera.
Response: EPA disagrees with the comment on cost. Please see response to comment EPA
HQ-OAR-2009-0923-0049-7 and the preamble Section III.B. EPA has revised todays final rule
to allow specific alternative leak detection methods. Please see the preamble Section II.F and
response to comment EPA-HQ-OAR-2009-0923-1039-18.
Commenter:
1123

The OVA/TVA requires the operator to physically access the emissions source with the probe
and thus is much more time intensive than using the optical gas imaging instrument. In addition,
the OVA/TVA range is limited to the reach of an operator standing on the ground or fixed
platform, thus excluding all emissions out of reach. However, EPA is seeking comments on
allowing the OVA/TVA to be used as another option to the optical imaging camera in this
proposed rule. (page 75)
MidAmerican supports allowing the use of OVA/TVA. Both OVA/TVA and optical gas imaging
instruments are capable of detecting leaks. There may be circumstances where one instrument is
preferred over another. MidAmerican believes that providing companies flexibility will produce
the most accurate results.
0923-1039-18.
If leak surveys are required, allow the use of OVAs and TVAs.
0923-1039-18.

Organization: Montana-Dakota Utilities Co.
Commenter: Abbie Krebsbach
The leak detection requirements in 98.233(q) and 98.234(a) of the Subpart W Rule state that
optical gas imaging instruments must be used to inspect for fugitive emissions annually at natural
gas distribution meter and regulator stations. Optical imaging instruments are expensive and cost
between $30,000 and $100,000 per device depending on what the EPA defines as an optical gas
imaging instrument. Many instruments would need to be purchased and additional staff hired to
conduct the required annual leak detection tests at all company meter and regulator stations. A
quick estimate indicated that each MDU LDC may need to potentially purchase 50 instruments
to conduct annual tests at all meter and regulator stations. Also, 20 to 30 additional employees
per MDU LDC may be needed to conduct the annual testing. The optical gas imaging
instruments and additional staffing would cost each MDU LDC approximately $6.5 million
which is too high a cost for our customers to bear, especially since there is no benefit.
1124
The cost to implement leak detection in this manner is excessively burdensome to the industry
and is not required by other programs. For instance, the DOT accepts soap solution leak
detection and leak detection wands for conducting leak inspections and less than annual
inspection frequencies. Again, MDU references AGA's comments which contain greater detail
on the projected costs and labor, as well as additional information AGA provides regarding
potential inaccuracy in measurement that occurs with optical imaging instruments due to wind
and interference from nearby emitting sources. Even if the EPA were to only include city gate
and district stations, MDU believes using optical imaging instruments for determining whether
fugitive emissions are occurring would still be too costly and not more accurate than other
approved leak detection methods.
Leak detection surveys as prescribed by DOT are not applicable to subpart W. Please see

OGI is the Most Effective System of Emissions Detection for Petroleum and Natural Gas
Systems
EPA has invited comment on allowing organic vapor analyzers (OVA) or toxic vapor
analyzers (TVA) to be used as an alternative to OGI for detecting fugitive methane emissions.
In our view, any leak detection program that does not incorporate OGI (or any comparably
effective leak detection technology) will miss large numbers of significant leaks and result in
reports that completely mischaracterize the magnitude and sources of GHG emissions.
The reason for this is quite simple: OVAs and TVAs (as well as the use of soap bubble
techniques) are only capable of detecting leaks when applied by a skilled operator at extremely
close range, along every square inch of a potentially leaking component. In our experience, these
methods are impractical for large, complex facilities such as petroleum and natural gas systems,
many of which have emitting components that cannot be safely or conveniently accessed by an
OVA/TVA operator.
1125
Since the time we built and placed our first OGI camera on the market, our customers have found
thousands of substantial leaks in petroleum and natural gas systems that were completely
overlooked by well-trained, well-equipped operators using OVA/TVA and soap bubble
inspection techniques. OGI technology found leaks on components that were not tagged or were
in areas of the facility considered too difficult to monitor under EPAs VOC regulations.
Inspectors equipped with OGI routinely find large leaks that are the largest contributors to a
facilitys overall emissions. Examples of emission sources frequently identified with OGI but not
included on the list of tagged components routinely inspected by OVA/TVA technology
include weld connections, cracked welds, and open-ended lines.
The link below illustrates a common occurrence whenever OGI technology is demonstrated to a
customer currently using TVAs. The video was collected with an OGI system in April of this
year on an offshore oil platform. The video clearly demonstrates how a trained technician
equipped with TVAs can easily miss a significant leak by placing the TVA probe in the wrong
area or upwind from the precise location of the leak source. On the leak survey depicted in the
video, two out of the three leaks identified by OGI were not detected with the companys TVAs.
http://www.flir.com/thermography/americas/us/OGI/offshoreplatforms/
In addition to being the most technically effective technology for detecting fugitive emissions,
OGI has other important benefits that EPA should consider. First, OGI enhances the safety of
petroleum and natural gas facilities by avoiding the need for operators to use OVAs/TVAs in
hazardous or inaccessible areas of facilities. In addition, OGI is capable of detecting potentially
explosive leaks that would otherwise pose serious safety risks at facilities.
Second, OGI can and does yield considerable economic savings for facilities that use the
technology to identify losses of natural gas. Unlike CO2, methane is a valuable product whose
loss in the form of fugitive emissions is a significant cost for natural gas facilities. One study
performed by Conoco Philips Canada at 9 gas plants and 13 compressor stations identified 144
leaking components that collectively accounted for 58.26 million cubic feet of methane per year,
or $358,012.10 USD/year in lost product. The company estimated that 92% of the 144 fugitive
sources are economical to repair, yielding savings with a net present value of $2,002,602.72
USD 136 FLIR Systems believes that such savings should be offset against the cost of OGI in the
Economic Impact Analysis accompanying the proposed Subpart W, and considered in EPAs
ultimate analysis of leak detection methods.
The above benefits have been quickly recognized by companies that have adopted OGI
technology. For example, an official from Shell has remarked:
[U]sing the camera for testing offers many advantages over more conventional technologies. The
camera is a quick, non-contact measuring instrument that can also be used in hard-to-access
locations. It also offers benefits in terms of safety and the environment - Rutger Zoutewelle,
136
(PILOT STUDY: Optical Leak Detection & Measurement, 2006).

1126
Shell Nederland Raffinaderij in Pernis (Rotterdam) 137 .

Unfortunately, despite the clear benefits of OGI, our experience has shown that allowing
facilities to choose between OGI and OVA/TVA techniques does not lead to sound regulatory
results. In practice, many petrochemical and petroleum refineries have opted to avoid OGI
entirely perhaps because of cost concerns or a fear of incurring penalties for finding previously
undiscovered leaks and instead use the technologies that miss large leaks from untagged and
uninspected components.
These components include equipment such as tanks (storage tank and condensate tank vents),
compressors (the compressor units themselves, as well as the compressor blow-down vents),
flares, and pressure relief valves. In addition, emissions from other components that are not
usually tested in leak detection and repair (LDAR) programs, such as weld connections, cracked
welds, and open-ended lines, can be the source of a facilitys largest leaks and will go undetected
because these areas are not on the list of tagged components.
0923-1039-18. Regarding the comment on costs, please see the preamble Section III.B and the
Economic Impact Analysis Section 5.

Organization: Dominion Resources Services. Inc.
Commenter: Pamela Faggert
The proposed rule requires an annual leak detection of fugitive sources for each of the source
categories listed in the proposed subpart. The prescribed methodology for performing these
annual detection surveys requires the use of optical gas imaging equipment. Due to the
combustible nature of natural gas, some locations within a facility may require equipment to be
intrinsically safe. Much of the optical gas imaging equipment currently in use is not classified as
intrinsically safe and therefore cannot be used within these areas for leak detection. Requiring
the use of this equipment will further reduce the commercial availability of the necessary
equipment, as facilities will need intrinsically safe equipment to complete the annual detection
on all components at a facility. Dominion requests that the agency consider the use of other
detection methodologies, such as organic vapor analyzers, which are already commonly used in
the industry.
0923-1039-18.
137
(Lahaut, Tracing gas leaks - Maintenance and safety problems highlighted, Maintenance Magazine, 2008).
1127

Commenter: Kim Dang
Fugitive Emissions Detection. The proposed Subpart W permits only one method of detecting
fugitive emissions from leaking components, storage tanks, and reciprocating compressor rod
packing cases: optical gas imaging used according to 40 C.F.R. SECTION 60.18(i)(1) and (2) of
the Alternative Work Practice. 138 EPA should not bind the petroleum and natural gas industry
to a single fugitive emissions detection technology. The inclusion of only optical gas imaging
would stifle innovation in emissions detection, and make the oil and gas industry captive to the
relatively small collection of manufacturers and contractors that produce and operate gas
imaging equipment. Indeed, Kinder Morgan and other members of our industry are concerned
that the proposed requirement would create an immediate surge in demand for optical gas
imaging services, which the industry may not be able to meet.
In addition, EPAs proposed optical imaging requirement would prevent members of our
industry from taking advantage of potentially more efficient and effective technologies that are
either close to being commercialized or could be developed in the future. For example, EPA
mentions laser technology (presumably the infrared laser leak detector) in the preamble to the
proposed rule, but does not appear to seriously consider this technique even though, as in optical
gas imaging, it permits remote detection of methane emissions within an operators line of
sight. 139,140 It is reasonable to expect that consensus standards for these new methods will soon
be developed, but only if EPA actually encourages the development of such standards by
allowing detection technologies other than optical gas imaging. Accordingly, Kinder Morgan
recommends that the proposed 40 C.F.R. SECTION 98.233(q) be modified to allow appropriate
standard methods of detection to be used including but not limited to optical gas imaging.
0923-1039-18.
Commenter: Kim Dang
EPA should allow the use of other standard methods for leak detection in addition to optical gas
imaging such as organic vapor analyzers, portable hydrocarbon gas detectors, ultrasonic meters
138
Proposed 40 C.F.R. SECTION 98.233(q).
139
Proposed Subpart W, 75 Fed. Reg. 18,623.] Partners in EPAs Natural Gas STAR program have also reported the
use of ultrasound techniques to detect gas escaping from valves and similar components.
140
Natural Gas STAR, Partner Reported Opportunities (PRO) Fact Sheet No. 602.
1128
and detectors and soap solution. Binding the petroleum and natural gas industry to a single
detection technology will stifle innovation.
0923-1039-18.
CAPP believes the prescriptive nature of 98.234(a) is unnecessarily burdensome and has chosen
a very specific instrument which is not widely used or accessible. There are a limited number of
consultants that offer this form of leak detection services. CAPP recommends that the EPA
allows other leak detection methods to be used for all source types listed in 98.233(p)(3)(i) and
(q) in operation or on standby mode that occur during a reporting period.
0923-1039-18.
Pre-amble-page 75, "EPA is seeking comments on allowing the OVA/TVA to be used as another
option to the optical imaging camera in this proposed rule."
- Since emission estimation from fugitive equipment leaks is based on identification of leaking
components and the use of appropriate leaker emission factors, CAPP supports the flexibility to
use OVA/TVA as another option to optical imaging cameras. If OVA/TVA were used, EPA
would have to adopt a leak definition consistent with that used to derive the leaker emission
factors (probably 10,000 ppm).
0923-1039-18. In addition, in todays final rule EPA specifies an instrument reading of 10,000
ppm or greater as a leak using Method 21 instruments. The Clearstone emission studies
referenced in the Technical Support Document (TSD) for todays final rule found in docket
(EPA-HQ-OAR-2009-0923); used 10,000 ppm or greater as a leak using Method 21 instruments.
1129

Optical Scanning/ Infrared Camera is Costly and Not Effective for Small Leaks; Therefore It
Should Not be Required
EPA explains in the preamble to the 2010 Proposal that EPA proposes conducting fugitive
emissions detection and then applying leaking component (or leak only) emissions factors for
processing, transmission, underground storage, LNG storage, LNG import and export terminals,
and LDC gate stations. 141 Proposed section 98.233(q) requires LDCs and other facilities subject
to Subpart W to use the methods described in 98.234(a) to conduct an annual leak detection of
fugitive emissions from the relevant list of components in 98.232. 142 Section 98.234(a) in turn
requires the use of an optical gas imaging instrument for fugitive emissions detection in
accordance with 40 CFR part 60, subpart A, 60.18(i)(1) and (2) Alternative work practice for
monitoring equipment leaks. There is no definition of optical gas imaging instrument in
proposed Subpart W. However, the referenced work practice in Part 60, defines the term in 40
C.F.R 60.18(g)(4) as:
(4) Optical gas imaging instrument means an instrument that makes visible emissions that may
otherwise be invisible to the naked eye.
Section 60.18(i)(1) and (20, as referenced in Subpart W proposed section 98.234(a) further
provides:
i) An owner or operator of an affected source who chooses to use the alternative work practice
must comply with the requirements of paragraphs (i)(1) through (i)(5) of this section.
(1) Instrument Specifications. The optical gas imaging instrument must comply with the
requirements in (i)(1)(i) and (i)(1)(ii) of this section.
(i) Provide the operator with an image of the potential leak points for each piece of equipment at
both the detection sensitivity level and within the distance used in the daily instrument check
described in paragraph (i)(2) of this section. The detection sensitivity level depends upon the
frequency at which leak monitoring is to be performed.
(ii) Provide a date and time stamp for video records of every monitoring event.
(2) Daily Instrument Check. On a daily basis, and prior to beginning any leak monitoring work,
test the optical gas imaging instrument at the mass flow rate determined in paragraph (i)(2)(i) of
this section in accordance with the procedure specified in paragraphs (i)(2)(ii) through (i)(2)(iv)
of this section for each camera configuration used during monitoring (for example, different
lenses used), unless an alternative method to demonstrate daily instrument checks has been
approved in accordance with paragraph (i)(2)(v) of this section...
141
75 Fed. Reg. at 18622.
142
75 Fed. Reg. at 18642.

1130
Although we are not entirely certain, we believe this means that the optical gas scanning
equipment required to be used for Subpart W leak surveys refers to what is known as an infrared
camera.
response to comment EPA-HQ-OAR-2009-0923-1039-18. In addition, the optical gas imaging
instrument is defined in the Alternative Work Practice to Method 21, and requires no further
definition in subpart W.
Allow Broader Range of Leak Detection Equipment--In the 2010 Proposal, EPA is apparently requiring the use of infrared (FLIR) cameras for the
annual leak surveys under Subpart W. This costly option is not necessarily the best option. In
fact, one of our member companies performed a leak survey using a FLIR camera recently to
assess whether it might improve leak detection for purposes of a potential project for the EPA
Natural Gas STAR program. First the LDC used normal, commonly used leak detection methods
in use for purposes of complying with existing federal and state leak survey requirements. Then
the LDC followed up with the FLIR camera to see if the camera could detect additional leaks
that were missed by the traditional leak detection equipment. The results were disappointing.
When used at a natural gas distribution city gate, the FLIR camera did not perform as EPA
assumes in the 2010 Proposal it could not scan hundreds of source types at the station quickly.
The FLIR did not find any additional leaks. In fact, the FLIR camera did not even find the small
leaks that the LDC had already located using normal leak detection methods even when one of
the field personnel pointed at the location of the leak so the person holding the camera could
point the FLIR directly at that location. They still could not see the leak with the FLIR. It appears
that the FLIR may be useful in some settings -- such as high pressure facilities where leaks tend
to be larger when they occur but it does not appear as useful for finding the small leaks in tight
LDC facilities.
In addition, EPA proposes to prohibit the use of Organic Vapor Analyzers (OVA). We disagree
with EPAs assumption about the accuracy of this equipment. Our members have demonstrated
that OVAs are a proven technology for which extensive operating procedures and trained
employees already exist.
Accordingly, to reduce costs and to improve results, AGA urges EPA not to require the use of
optical gas scanning equipment/ infrared cameras, and instead to allow the use of a full range of
available leak detection equipment, allowing the operator to select the tool that is most effective
for a given situation. The best way to do this may be to reference existing federal and state leak
1131
detection regulations and relevant portions of the GPTC industry standard guidance for leak
surveys.
response to comment EPA-HQ-OAR-2009-0923-1039-18. EPA disagrees that optical gas
imaging instruments are inferior to alternative leak detection methods. Please see EPAs
analysis of the optical gas imaging instrument in EPAs preamble to the Alternative Work
Practice to Method 21, and in docket EPAHQ-OAR-2003-0199-0005.
Allow Broader Range of Leak Detection Equipment for Leak Surveys at Underground Storage
Facilities
Similarly, EPA should not require the use of optical gas scanning equipment (infrared FLIR
cameras) but should allow a broader range of effective leak detection equipment for leak surveys
at underground storage facilities, as we argue elsewhere in these comments.
0923-1039-18.
Section 98.233(q) Leak detection and leaker emission factors
Currently, Section 98.233(q) requires surveying standard components (such as connectors and
valves, as opposed to vented components) with an IR camera and then applying a leaker
emission factor to the number of leaks found for each component type. Its clear that EPAs
intent was to reduce labor costs in this area. However, this approach still requires a substantial
amount of effort that does not provide a large improvement in the accuracy of the reported
emissions. It would also distract from collecting accurate data from the vented components that
make up the majority of non-combustion emissions.
The contribution of leakage from standard components averages about 20% of the total leakage
at transmission compressor stations (Howard et al., 1999 this was a joint study sponsored by
1132
PRCI, GRI, and EPA) 143 . Because the proposed method uses an emission factor as opposed to a
measurement, there will still be uncertainty in this method. The EPA Natural Gas Star program
has previously developed emission factors for compressor stations based on the number of
compressor units at the site (EPA, 2000) 144 . These factors include both standard and vented
components but could easily be recalculated to assess only standard components.
Although the proposed method of using leak detection combined with leaker emission factors
would be expected to give better results than applying a facility emission factor, this
improvement only applies to approximately 20% of the total facility leak rate. Consequently, the
improvement in the reporting accuracy for the entire facility is probably less than 5%, while the
effort involved might require as much as 50% of the total leak survey effort.
Allowing a facility emission factor for standard components as an option to leak screening lets
facilities focus on making measurements at the vented components that account for 80% of the
total emissions. This full focus should improve the accuracy of the data collected from these key
sources.
Should EPA still feel that leak surveys of standard components are essential, it would be better to
allow more options than just the IR camera to conduct these surveys. Although these cameras are
useful tools, screening with flame ionization detectors, catalytic oxidation/thermal conductivity
detectors, and soap solutions have been proven to be effective (Howard et al., 1999; EPA, 2002;
KSU, 2006) and should not be excluded.
Its possible that EPA felt that the IR cameras would be the most cost effective approach, and for
some companies, that may be true. However, other companies might want to have in-house
personnel conduct leak screening with techniques that they already know how to use.
Clearstones 2006 study (KSU, 2006) 145 notes that a 2 person crew with an IR camera can survey
approximately 6,400 components per day. The PRCI study (p. A-1-14) found that experienced
personnel could survey transmission station components using a combination of soap and
catalytic oxidation/thermal conductivity detectors at a rate of 350 components per person per
hour, or 5,600 components per two person crew per eight hour day. During this time study, the
personnel found 99.6% of combined leaks found by using this method and a Foxboro TVA
1,000, so these methods are clearly still effective.
Facilities that have access to IR cameras and trained personnel could apply the IR camera if they
so desired. Other facilities without these resources could rely on past experience and conduct the
leak screening on a flexible time schedule without necessarily acquiring more equipment or
143
Howard, T., R. Kantamaneni, and G. Jones, 1999. Cost Effective Leak Mitigation at Natural Gas Transmission
Compressor Stations. PR-246-9526. PRCI, 1515 Wilson Boulevard, Arlington, VA22209.
144
EPA, 2000. Natural Gas STAR Default Value Analysis Project. 7th Annual Natural Gas STAR Implementation
Workshop, October, 2000.
145
National Gas Machinery Laboratory, Kansas State University; Clearstone Engineering, Ltd; Innovative
Environmental Solutions, Inc., 2006. Cost-Effective Directed Inspection and Maintenance Control Opportunities
at Five Gas Processing Plants and Upstream Gathering Compressor Stations and Well Sites. For EPA Natural Gas
STAR Program. March 2006.
1133
hiring contractors.
Contribution of Standard Components to Total Leakage
There may be an overemphasis on the importance of standard components at facilities due to
Clearstone Engineerings studies at gas plants in 2002 and 2006. Although these studies were
detailed and well executed, gas processing plants have a much larger ratio of standard
components per compressor unit than is found at transmission compressor stations. For instance,
in the PRCI study done at thirteen transmission compressor stations, the number of components
per compressor unit was a little over 400. The Clearstone studies dont list a complete inventory
of compressor units, but if there were twenty units at each site, the ratio of components to
compressor units would be approximately 1,000, about two and one half times greater than at
transmission sites. Additionally, standard components at processing plants may be subjected to
greater heat and vibration in some areas than at transmission compressor stations.
The 2002 Clearstone study (EPA, 2002) 146 indicated that connectors and valves accounted for
over 50% of the total leakage (Figure 10 of that study), whereas at transmission stations that
number is more likely to be 20% (Howard et al, 1999). Additionally, the 2006 Clearstone study
(KSU, 2006) indicates that connectors and valves were down to approximately 32% (Figure 13
of that study). Open ended lines contributed another 30%, but this category apparently includes
compressor blow down valve leakage which would probably dominate the leakage.
It is understood that for the purposes of the proposed rule that different leaker emission factors
have been developed for transmission as opposed to processing. However, the key point is that
standard components in transmission dont contribute to the total leakage as much as the
Clearstone studies might be interpreted to suggest.
Response: Please see response to comments EPA-HQ-OAR-2009-0923-0055-8 and EPA-HQ
OAR-2009-0923-0055-14.
Organization: American Public Gas Association (APGA)
Commenter: Bert Kalisch
The Final Rule should allow other leak detection equipment for LDCs in addition to optical gas
imaging equipment.
By operation of Section 98.233 (q), all LDC facilities identified by Section 98.232(i)(1) are
required to be leak surveyed annually. Section 98.234(a)(1) requires that such surveys use an
optical gas imaging instrument. APGA appreciates that EPAs intent is to allow LDC operators
to use equipment that can detect and count leaking components from a distance and thus reduce
146
EPA, 2002. Identification and Evaluation of Opportunities to Reduce Methane Losses at Four Gas Processing
Plants. Clearstone Engineering Ltd. June 20, 2002. www.epa.gov/gasstar/documents/four_plants.pdf.
1134
the burden of conducting these surveys. APGA supports allowing utilities to use this method, but
does not support requiring its usage. EPA should allow LDCs to use the equipment which they
currently use for leakage surveys and all of which are at least (and in many cases are much more)
accurate than the optical imaging equipment that would be required by the Proposed Rule. In
addition to optical gas imaging equipment, leakage surveying equipment used by gas distribution
operators includes:
Flame-ionization Detector (FID)
Catalytic Combustion / Thermal Conductivity Combustible Gas Indicator (CGI)
Optical
Laser
APGA stresses that the cost of purchasing an optical gas imaging device or hiring a contractor to
leak survey with such a device would be entirely unnecessary as an optical imaging device
would provide no better results that the leak detection equipment listed above, at least one of
which LDCs already own and use effectively. Moreover, optical imaging devices are much more
costly than many of the leakage detection equipment listed above.
We do not see how the requirement to limit leak detection technologies solely to optical gas
imaging instruments can be justified, at least in the LDC context. This overly restrictive
requirement is a disincentive for further innovation, and in the case of LDC's, requires additional
costs and training that cannot be economically justified. As the Technical Support Document
(TSD) (p. 25) notes, "traditional technologies like Toxic Vapor Analyzer (TVA) and Organic
Vapor Analyzer (OVA) are appropriate for use in small facilities with few pieces of equipment".
The TSD further notes that "the main disadvantage of an IR camera is that it involves substantial
capital investment ... therefore, these cameras are most applicable in facilities with large number
of equipment and multiple potential leak sources or when purchased at the corporate level, and
then shared among the facilities."
While an LDC might "share" a single optical gas imaging device as the TSD notes, it is still
questionable whether this would truly reduce costs once the need for specially trained staff and
travel costs to survey geographically dispersed sites are factored in. In the Canadian context, an
LDC's facilities can span significant areas within a province. On the other hand, company
1135
employees within each region can use existing, on-hand leak detection equipment such as OVA's
to achieve the same objective of leak detection.
We therefore suggest that EPA remove the equipment specification related to leak detection,
leaving the choice of technology to reporting companies to choose the monitoring method most
appropriate to their operations.
OVA/TVA Monitoring
EPA is seeking comments on allowing the OVA/TVA to be used as another option to the optical
imaging camera in this proposed rule.
BP supports the use of OVA/TVA as an alternative option to the optical imaging camera
particularly where a current LDAR program is in place under Subpart KKK.
0923-1039-18.
Alternatives to the proposed methods for leak detection should be considered to provide more
flexibility in meeting the requirements of the proposed rule
EPA should allow alternatives to the optical gas imaging instrument described in Sec.
98.234(a) for detecting fugitive emissions from each source category. Limiting the fugitive
emissions leak detection method to one technology does not recognize the numerous tools that
are already used by the industry to detect fugitive leaks and will potentially create shortages of
instruments or trained technicians in the first year or two of the program. Restricting the types of
leak detection that can be used does not allow for technology improvements that could be made
to existing gas detectors and does not take into account the limited number of vendors that
provide this product or its high cost. EPA indicates that contractors could be used to do this leak
detection, and it is our understanding that a very limited number of contractors are available with
1136
optical gas imaging instruments. DTE Energy believes that EPA should provide maximum
flexibility in the rule for allowing the use of existing or yet to be developed methods for fugitive
leak detection. We support the use of IR Laser Detectors, Organic Vapor Analyzers (OVAs) and
Toxic Vapor Analyzers (TVAs) for fugitive leak detection.
Flexibility in evaluating fugitive leaks should also be provided for detecting leaks from
inaccessible or out of reach sources, including situations where safe working conditions cannot
be met by using an approved leak detection device or method. DTE Energy requests that EPA
allow for the use of population emission factors or use of manufacturers specification in these
special situations.
In addition, we are concerned about the lack of flexibility allowed for leak detection methods in
the proposed rule. The remainder of this letter explains our concerns in more detail.
0923-1039-18.
Alternatives to the proposed methods for leak detection should be considered to provide more
flexibility in meeting the requirements of the proposed rule
0923-1039-18.
Organization:
Commenter: Michael Webb
Locating Fugitive Emissions (98.234 Monitoring and QA/QC requirements)
1137
I propose that 98.234 (a) (2) be worded:

Method 21 approved methodologies or other monitoring techniques with a methane and/or CO2
minimum sensitivity limits equal to or less than the lowest value given in Table 1 to Subpart A of
Part 60 -Detection Sensitivity Levels (grams per hour) for the Optical Gas Imaging Instrument.
There are several methodologies current in use by the oil & gas and chemical industries that
locate very small leaks (such as sonic testing, volumetric displacement, etc.). 40 CFR Part 98
would be stronger if it included all of these proven methods.
Allowing the use of other methods that are at least as sensitive as the Optical Gas Imaging
Instrument will be technology-forcing. It will motivate companies to develop new equipment
now and in the future that can detect low emissions of methane and/or CO2.
For example, the Bacharach, Inc. Hi Flow Sampler, referenced in Appendix K of the
Background Technical Support Document for Subpart W, lists a minimum sensitivity level of
0.05 scf/min. This is approximately 57 grams/hr of methane. Therefore, the Hi Flow Sampler
might be useful for monitoring as well as quantifying leaks.
In the early 1990s, I developed an experimental quantification system, the High Volume
Collection System (HVCS), which was tested by the EPA both in the laboratory and in the field
in 1994. EPA reported on the success of the HVCS in Document EPA-600/R-95-167, Evaluation
of the High Volume Collection System (HVCS) for Quantifying Fugitive Organic Vapor Leaks.
A consumer version of the HVCS will be available later this year. It will locate and quantify both
carbon dioxide and hydrocarbon fugitive emissions. The minimum detection limits for methane
and CO2 will be less than 1 gram/hr. The new version will weigh 3 pounds, will be fully
automatic, and can be operated by anyone after a short training session.
As written, 98.234 (a) will prohibit the use of this and other new technologies. I recommend
that 98.234 (b) include language that will allow new technologies to be used.
0923-1039-18.
Organization:
In 98.234 (a) (2), allow the use of other techniques for locating fugitive leaks such as: Method
21 approved methodologies or other monitoring techniques with a methane and/or CO2
minimum sensitivity limits equal to or less than the lowest value given in Table 1 to Subpart A of
Part 60 -Detection Sensitivity Levels (grams per hour) for the Optical Gas Imaging Instrument.
1138
0923-1039-18.
Costly and Ineffective Technology
In this Proposed Rule, Southwest is concerned that EPA has not appropriately considered the
costs associated with gathering proposed information, let alone considered whether or not the
technology and/or information on which estimated emission rates is based is anywhere near
reasonably accurate.
Technology that is proposed to be used, such as IR laser detector instruments, Toxic Vapor
Analyzers (TVA) or Organic Vapor Analyzers (OVA) may have applicability in the future, but
they cost approximately $80,000 each. Southwest estimates that it would have to purchase at
least 25 of them in order to complete leak surveys within the two month time period envisioned
by the new rule. Even if there were enough instruments manufactured in time to comply with the
initial reporting deadline, it is unlikely that they could be deployed in the field in any reasonable
time frame to survey and collect information on literally thousands of possible, but unlikely,
emission sources.
In addition, Southwests experience with these instruments is that while the technology works for
large leaks, Southwest employees had to identify smaller leaks by traditional methods, such as a
soap solution, before the instrument operator could recognize the leak through the instrument
viewer. These instruments also do nothing to quantify any leakage found
Response: EPA disagrees with the comment on costs. Please see response to comments EPA
HQ-OAR-2009-0923-1020-6 and EPA-HQ-OAR-2009-0923-0049-7, and the preamble Section
III.B. EPA has revised todays final rule to allow specific alternative leak detection methods.
18. EPA disagrees that optical gas imaging instruments are inferior to alternative leak detection
methods. Please see EPAs analysis of the optical gas imaging instrument in EPAs preamble to
the Alternative Work Practice to Method 21, and in docket EPAHQ-OAR-2003-0199-0005.
Costly Ineffective Technology
1139
Southwest highly recommends that the regulations allow the use of industry standard leak
detection methods that have been demonstrated to be accurate, highly effective and efficient over
the past decades.
Optical scanning technology that is proposed to be used may have applicability in the future, but
they cost approximately $80,000 each and Southwest alone would have to purchase 25 of them
in order to complete the leak survey within the two month time period envisioned by the new
rule. Even if there were enough instruments manufactured in time to comply with the initial
reporting deadline, it is unlikely that they could be deployed in the field in any reasonable time
frame to survey and collect information on literally thousands of possible emission sources.
In addition, Southwests experience with these instruments is that while the technology works for
large leaks, Southwest employees had to identify smaller leaks with a soap solution before the
instrument operator could recognize the leak through the instrument viewer. These instruments
also do nothing in terms of volume released or even atmospheric concentration to quantify any
leakage found.
Comment Number: EPA-HQ-OAR-2009-0923-0053 -6
Additionally, more flexibility regarding the integration with existing leak detection and repair
(LDAR) programs would be appreciated. Most facilities currently subject to LDAR regulations
utilize OVA methods. Please consider allowing a combination of methods such as OVA/TVA for
components that are already being monitored through these methods and optical gas imaging for
other components.
Please consider allowing multiple alternate methods such as a "bubble test" method . This
method would be much more cost-effective and provide the same qualitative information as
compared to the optical gas imaging method.
Please consider allowing facilities to use alternate methods (other than optical gas imaging) in
the event that the facilities are capable of monitoring all components (including components that
may be considered "difficult to monitor" and are currently not being monitored as part of the
existing facility LDAR program). When faced with the high cost of optical gas imaging, facilities
may prefer to erect scaffolding or use fall protection harnesses or otherwise safely access
"difficult to monitor" components. The added benefit of facilities choosing this option is that
while personnel are accessing the component, some leaks may be repaired during the inspection
process.
1140
Section 98.234(b) Monitoring and QA/QC requirements
Subpart W Section 98.234(b) does not reference Section 98.3(i). Do the calibration and accuracy
requirements in Subpart A - General Provisions Section 98.3(i) apply to Subpart W; including
but not limited to the 5% accuracy and delay of calibrations until the next scheduled shutdown if
a shutdown is needed to conduct the calibration? (The monitoring and QA/QC requirements in
other subparts such as Subpart C - Combustion Section 98.34(b) and Subpart Y - Petroleum
Refineries Section 98.254(b) reference Section 98.3(i).)
Section 98.234(b) Monitoring and QA/QC requirements
Applicability of Section 98.234(b) - Paragraph 98.234(b) provides the monitoring and QA/QC
requirements for flow meters, composition analyzers and pressure gauges. For monitoring
methods in Section 98.233 for emissions sources where EPA specifies in Table W-4 of the
preamble (75 FR 18619) the monitoring method is direct measurement, Section 98.234(b) is
referenced. For example the flow monitoring method for transmission tanks, centrifugal
compressor wet seal degassing vents, and reciprocating compressor rod packing vents in Section
Section 98.233(k)(2)(i), 98.233(o)(1) and 98.233(p)(2)(ii), respectively, all reference Section
98.234(b). Conversely, the flow or volume value used to calculate emissions for gas well venting
during conventional well completions and workovers, well testing venting and flaring, and
associated gas venting and flaring in Section 98.233(h), 98.233(l) and 98.233(m), respectively,
do not reference Section 98.234(b). Thus, BP interprets Section 98.233(h), 98.233(l) and
98.233(m) to mean that the flowrate or volume values can be determined from flow meters that
do not meet Section 98.234(b), engineering estimates or company records. EPA should confirm
that the monitoring and QA/QC requirements in Section 98.234(b) do not apply to flow rate,
pressure, and composition values when Section 98.234(b) is not specifically referenced in
Section 98.233(a) through (z).
Response: Subpart A, and its instrument calibration procedures, applies to subpart W unless
otherwise specified in subpart W. Please see The Final Mandatory GHG Reporting Rule (Final
MRR), (40 CFR part 98) preamble Section L. EPA has revised todays final rule Section
98.234(b) to clarify this. EPA has proposed amendments to subpart A calibration requirements,
please see the MRR proposal August 2010.

1141

Section 98.234(b) Monitoring and QA/QC requirements. Subpart W Section 98.234(b) does not
reference Section 98.3(i). Do the calibration and accuracy requirements in Subpart A - General
Provisions Section 98.3(i) apply to Subpart W; including but not limited to the 5% accuracy and
delay of calibrations until the next scheduled shutdown if a shutdown is needed to conduct the
calibration? (The monitoring and QA/QC requirements in other subparts such as Subpart C Combustion Section 98.34(b) and Subpart Y - Petroleum Refineries Section 98.254(b) reference
Section 98.3(i).) API interprets Section 98.234(b) to mean that Section 98.3(i) does not apply to
Subpart W. If this interpretation is correct, API recommends EPA add the requirements in
Section 98.3(i)(5) and (6) regarding previous calibrations that are still active and deferring
calibrations to the next scheduled shutdown if a shutdown is needed to conduct the calibration
into Subpart W Section 98.234(b).

Section 98.234(b) Monitoring and QA/QC requirements. Applicability of Section 98.234(b) Paragraph 98.234(b) provides the monitoring and QA/QC requirements for flow meters,
composition analyzers and pressure gauges. For monitoring methods in Section 98.233 for
emission sources where EPA specifies in Table W-4 of the preamble (75 FR 18619) the
monitoring method is direct measurement, Section 98.234(b) is referenced. For example the flow
monitoring method for transmission tanks, centrifugal compressor wet seal degassing vents, and
reciprocating compressor rod packing vents in Section 98.233(k)(2)(i), 98.233(o)(1) and
98.233(p)(2)(ii), respectively, all reference Section 98.234(b). Conversely, the flow or volume
value used to calculate emissions for gas well venting during conventional well completions and
workovers, well testing venting and flaring, and associated gas venting and flaring in Section
98.233(h), 98.233(l) and 98.233(m), respectively, do not reference Section 98.234(b). Thus, API
interprets Section Section 98.233(h), 98.233(l) and 98.233(m) to mean that the flowrate or
volume values can be determined from flow meters that do not meet Section 98.234(b),
engineering estimates or company records. API requests confirmation from EPA that the
monitoring and QA/QC requirements in Section 98.234(b) do not apply to flow rate, pressure,
and composition values when Section 98.234(b) is not specifically referenced in Section
98.233(a) through (z).
Response: The monitoring and QA/QC requirements detailed in Section 98.234(b) apply to all
monitoring requirements in the rule, unless otherwise specified in subpart W.
1142

98.234 (b):
CAPP does not support the requirement for the re-calibration of ALL meters, analyzers and
pressure gauges annually. Although CAPP agrees that calibration is necessary, the frequency of
calibration should be based on manufacturers recommendations. Scheduling of required
calibrations should be left to the operator, since coordination with plant shutdowns might be
required.
Response: All aspects of Subpart A, including instrument calibration procedures, apply to
subpart W unless otherwise specified in subpart W. Please see The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98) preamble Section L EPA has revised todays
final rule Section 98.234(b) to clarify this.
Organization: United States Department of the Interior
Commenter: Willie R. Taylor
Page 18645, 98.234. Monitoring and QC/QA requirements. paragraph (b): This requires all
flow meters to use methods and calibration practices published by consensus organizations (e.g.
ANSI and API). It further states that if a consensus based standard is not available, the
manufacture's instructions must be used to calibrate a meter. This regulation may conflict with
the recently approved MMS regulations in 30 CFR 250, Subpart K. In Subpart K, MMS requires
operators to calibrate flare meters per the manufacture's recommendation or at least once per
year, whichever is shorter. ANSI, API, etc. recommendations could require a shorter, or allow a
longer, calibration frequency than Subpart K. Also, each time a consensus organization revises a
recommended standard, conflicts with Subpart K requirements would need to be assessed. We
recommend adding an additional sentence to paragraph (b) that would require offshore Outer
Continental Shelf operators to install, maintain and calibrate flare/vent meters in accordance with
30 CFR 250, Subpart K.
We did not see any mention of flare/vent meter accuracy requirements in this proposed rule. By
adding the above recommended sentence to 98.234(b), however, the accuracy and calibration
requirements for meters installed on facilities in federal waters would be consistent with MMS
requirements (30 CFR 250, Subpart K requires 5% accuracy).
Response: Regarding offshore production methodologies under subpart W, please see response
to comment EMAIL-0010-1 (comment also located in rulemaking memo Early Comment
Submissions in docket EPA-HQ-OAR-2009-0923).
1143
Monitoring and QA/QC requirements at 40 CFR 98.234:

As stated in comment number 2 above, Laclede suggests that programs currently in place at
LDCs to minimize the leakage of natural gas already provide strict, functional regulation in this
area, and we recommend EPA abandon this course altogether. If EPA does proceed with
requiring the counting of LDC components and annual leak surveys, EPA should ensure that
subpart W expressly recognizes and accepts the use of all standard leak detection methods,
technologies, and instruments approved now or in the future by PHMSA/DOT pipeline safety
regulations for LDCs to identify leaks at above and below ground M&R stations, valves,
compressor stations, wellheads, etc. The proposed rule is far too prescriptive with respect to the
leak detection methods and equipment allowed, and seems to indicate insufficient understanding
on EPAs part of the wide array of cost effective techniques at the natural gas distribution
industrys disposal for locating and identifying leaks.
Response: EPA disagrees that the rule is duplicative of DOT regulations. Please see response to
comment EPA-HQ-OAR-2009-0923-0955-7. EPA has revised todays final rule to allow
specific alternative leak detection methods. Please see the preamble Section II.F and response to

The proposed rule requires significant additional effort to log and report leaks beyond current
requirements for detecting and fixing leaks on distribution systems and does not recognize that
existing programs already minimize fugitive emissions from LDCs
Current regulations promulgated by the Department of Transportation (DOT) Pipeline and
Hazardous Substances Administration (PHMSA) under Title 49 of the U.S. Code of Federal
Regulations (CFR) and the Michigan Gas Safety Standards require utilities to perform leakage
surveys on distribution systems and to follow up with leak repair on a schedule that is
appropriate to the potential hazard of the leak to the public or to buildings. The implementation
of this existing compliance program continually identifies and reduces emissions of GHGs from
natural gas distribution systems.
In addition, a final order from the Michigan Public Service Commission (MPSC) to MichCon
requires MichCon to prepare and submit a long term plan to significantly reduce the amount of
cast iron main in MichCons distribution system. The MPSC order, dated June 3, 2010, was part
of a response to a rate case filing. Since cast iron natural gas piping has the highest emission
factor for leaks among the various gas main construction materials, this order will have the effect
of significantly reducing fugitive emissions of GHGs from MichCons distribution system.
DTE Energy is concerned that the leak detection requirements for distribution systems under
proposed Subpart W will add significant additional regulatory burden to a leak detection
1144
program that is already successful in identifying and fixing leaks. The proposed rule would
require that all leaks detected will be recorded into a centralized system to be able to be reported
to EPA. If optical imaging is used as the detection method, then the imaging video will be used
as the record and identified leaks will be input directly into a database by the technician
reviewing the camera footage. Technicians will require training to comply with EPA rules and in
properly citing and reporting all leaks detected and/or repaired. DTE Energy does not currently
employ such a system for leak detection and repair and would require significant time to adjust
to such a procedure. If a rule is finalized in late 2010, only a few months will be available to
implement the full procedure for performing inspections and documenting leaks identified by
current procedures.
comment EPA-HQ-OAR-2009-0923-0955-7. EPA has revised todays final rule to allow
specific alternative leak detection methods. Please see the preamble Section II.F and response to
comment EPA-HQ-OAR-2009-0923-1039-18. EPA has revised the rule and videos are not
required. Please see response to comment EPA-HQ-OAR-2009-0923-1024-49.

Furthermore, LNG storage facilities, under the US Department of Transportation (DOT) Pipeline
and Hazardous Materials Safety Administration (PHMSA) are required to have leak and
flammable gas detection systems. The PHMSA also mandates the monitoring of such systems
and the repair of any leaking component or equipment. Also, LNG storage facilities typically
vaporize fewer than 15 times annually during times of cold-weather peak gas demand. Such
infrequent operations limit the probability and amount of fugitive emissions.
comment EPA-HQ-OAR-2009-0923-0955-7. EPA disagrees with the comment on LNG
facilities, and is including the LNG facilities in todays final rule. Please see the response to

EPAs proposed rule would require natural gas LDCs to report fugitive emissions from the
following sources: 1) above grade M&R and gate stations, 2) below grade M&R stations and
vaults, 3) pipeline mains, and 4) service lines. The Clean Energy Group generally agrees with the
sources identified by EPA for inclusion in the emission reporting program. However, the
proposed rule provides no differentiation regarding the size, pressure, or purpose of above grade
1145
M&R stations, so it is unclear which stations should be surveyed. The Clean Energy Group
contends that emission reporting from these sources should be based on physical attributes such
as size, pressure, or capacity, and that customer meters should not be included, with the possible
exception of high capacity industrial meters, as this would make the process unmanageable due
to the sheer number of customer meters.
The Clean Energy Group also finds that the requirement to annually survey M&R stations using
an optical gas imaging instrument (i.e., camera) is unreasonable. It is unclear what benefit if any
this technology provides relative to other methods more widely used by the industry to detect
leaking equipment that are far more cost effective. Furthermore, most LDCs do not own these
cameras and have no experience using them. As a result, companies would have to purchase
these cameras, develop procedures and train staff to conduct these annual surveys and/or pay
outside firms to conduct the surveys.
The Clean Energy Group believes the costs associated with this requirement renders it infeasible,
as detailed below, and recommends that EPA allow the use of population emission factors to
estimate the emissions from this source, or to allow leak surveys to be conducted less frequently
perhaps in alignment with current leak survey requirements specified by 49 CFR 192.723
Distribution systems: Leakage surveys.
Also, many LDCs operate a large number of small to very small distribution sub-systems in rural
and remote areas in addition to large integrated distribution sub-systems in metropolitan areas.
The costs for special leak surveys of these small, remote sub-systems are extremely high relative
to the benefits of conducting them. If the use of population emission factors to report emissions
from above-ground M&R stations is not allowed, the Clean Energy Group suggests that
distribution facilities be redefined to exclude small, remote sub-systems in order to reduce
reporting costs. Alternatively, the rule should include language that addresses the very high cost
of special surveys for small, remote sub-systems.
Response: Todays final rule clarifies that only aboveground M&R city gate stations at custody
transfer require leak detection. Please see the preamble Section II.F. EPA disagrees with the
comment on DOT regulations. Please see response to comment EPA-HQ-OAR-2009-0923
0955-7. EPA disagrees with the comment on cost. Please see response to comment EPA-HQ
OAR-2009-0923-0049-7 and the preamble Section III.B. EPA has revised todays final rule to
allow specific alternative leak detection methods. Please see the preamble Section II.F and

Organization: Xcel Energy Inc.
Commenter: Eldon Lindt
The proposed rule will place limitations on the leak survey technology used for annual leak
surveys in city gate stations, metering and regulating (M&R) stations, valve sets, processing
plants, storage fields and compressor stations. The proposed technology may not be the best or
1146
most effective equipment for a particular situation. The proposed requirements also will require
the LDCs to perform additional leak surveys, above those required by DOT Code. Performing
these annual leak detection surveys using only the specified technology would likely overwhelm
the market for such equipment and strain company resources since LDCs could be performing
leak surveys on millions of stations. Xcel Energy supports allowing LDCs to use any leak
detection equipment accepted by current industrial standards and/or practices. At the very least,
use of Best Available Monitoring Methods (BAMM) should be allowed for the first reporting
year.
response to comment EPA-HQ-OAR-2009-0923-1039-18. Concerning the availability of
equipment, for certain sources EPA will allow the application for the use of best available
monitoring methods. Please see the preamble Section II.F.
Fourth, EPA should delete the requirement to conduct annual leak surveys at city gates and
above ground M&R stations and the requirement to use infrared cameras or other types of optical
gas scanning equipment. This duplicates existing leak survey requirements imposed at the
federal and state level by DOT PHMSA and state utility commissions. In addition, leak surveys
are unnecessary for GHG estimation purposes if EPA uses a facility level rather than a
component level - emission factor for city gates and M&R stations.
Response: Todays final rule clarifies that only above ground M&R city gate stations at custody
transfer require leak detection. Please see the preamble Section II.F. EPA disagrees with the
comment on DOT regulations. Please see response to comment EPA-HQ-OAR-2009-0923
0955-7. EPA has revised todays final rule to allow specific alternative leak detection methods.
18. Regarding the comment on facility level emission factors for M&R stations, please see

Proposed Leak Surveys with Optical Scanning Are Not Necessary or Useful; They Duplicate
Current Federal/ State Requirements
There is also no need for the annual leak surveys called for under Subpart W. EPA proposed
1147
these leak surveys in order to identify leaking components in city gate stations and above ground
M&R stations. The proposed rule then would require LDCs to apply special emission factors for
leaking components to calculate fugitive methane emissions. There are several problems with
this proposal. First, infrared cameras are often not the best choice for locating leaks in
distribution facilities. Second, infrared cameras are far more expensive than other reliable,
effective leak detection methods. They cost between $80,000 and $100,000 each, compared to
other options that cost as little as a few hundred dollars each. Third, federal and state agencies
with many years of experience and expertise already require LDCs to perform regular leak
surveys. EPAs proposal would duplicate and potentially conflict with those federal and state
leak detection requirements.
Response: EPA disagrees with the comment on DOT regulations. Please see response to
comment EPA-HQ-OAR-2009-0923-0955-7. EPA disagrees that optical gas imaging
instruments are less useful than alternative leak detection methods. Please see EPAs analysis of
the optical gas imaging instrument in EPAs preamble to the Alternative Work Practice to
Method 21, and in docket EPAHQ-OAR-2003-0199-0005. EPA disagrees with the comment
on cost. Please see response to comment EPA-HQ-OAR-2009-0923-0049-7 and the preamble
Section III.B. EPA has revised todays final rule to allow specific alternative leak detection
0923-1039-18.

DOTs Pipeline and Hazardous Substance Safety Administration PHMSA already requires
utilities to conduct leak surveys of gas distribution systems:
1. Once per year in business districts annually and
2. Once every 5 years outside business districts.
The PHMSA natural gas pipeline safety rules direct LDCs to use any effective leak detector
equipment. See 49 C.F.R. 192.723. This rule allows LDCs to use the best, most effective
equipment for the particular situation, choosing from the full range of available options, such as
soap solution, gas detection wands, and various types of optical scanning equipment. This
approach is less burdensome and frankly more effective than EPAs duplicative proposal to
require the use of optical scanning equipment which we believe refers to infrared cameras. Our
members have conducted field tests recently to compare the effectiveness of infrared cameras
and other types of leak detection. They found infrared cameras (the most expensive of these
options) can be useful for some purposes, but the cameras often fail to detect leaks in natural gas
distribution facilities even after the operator has already located the leaks using less costly,
normal methods for leak detection such as soap solution and leak detection wands.
1148
In addition, while some states adopt the PHMSA leak detection requirements by reference, many
state public utility commissions (PUCs) impose more stringent standards for leak survey
frequency and/or specify a range of acceptable, alternative leak detection methods. Some states
require LDCs to conduct annual leak surveys of all distribution facilities regardless of where they
are located, whereas PHMSA allows less frequent surveys for facilities located outside business
districts.
Finally, there is no need to apply emission factors for leaking components to arrive at an
estimate of methane emissions from city gate stations and above ground M&R stations. Subpart
W proposes to apply a facility-level emission factor to below ground M&R stations. EPA has
provided no reason for not using a similar approach for above ground M&R stations and city
gates. This would eliminate the need to visit hundreds of thousands of industrial and commercial
customer meters and regulators or millions of residential meter & regulator sets to count
specific types of components (such as couplings and valves) or to perform duplicative leak
detection using scarce and expensive infrared cameras.
Response: Please see response to comment EPA-HQ-OAR-2009-0923-0049-5. Regarding the
comment on facility level emission factors for M&R stations, please see response to comment
EPA-HQ-OAR-2009-0923-1065-4. EPAs analysis of using population emission factors for
below ground M&R stations is in the April 2010 preamble Section II.E. EPA has revised todays
final rule to allow specific alternative leak detection methods. Please see the preamble Section
II.F and response to comment EPA-HQ-OAR-2009-0923-1039-18

PHMSAs regulations, which are found in Title 49 Code of Federal Regulations (49 CFR) Parts
192 and 193, establish the minimum requirements that all pipeline operators must meet. Notably,
the leak survey provisions of 49 CFR 192.721 and 192.723 and the gas and fire detection
requirements of 49 CFR Part 193 apply. It is Southwests opinion that EPA apparently has not
considered the above referenced rigorous pipeline safety regulations under which both interstate
and intrastate pipelines operate with regard to leak detection. 147 In so doing EPA appears to be
ignoring the records of stringent leak survey and repair procedures that have been used by gas
transmission, distribution and LNG operators for over 50 years, and for over 30 years as required
by federal and state pipeline safety regulations, all of which demonstrate that gas pipelines use
pipeline components that contain their product very well
Response: EPA disagrees with the comment on DOT regulations. Please see response to
147
For an example of state regulations, see Arizona Administrative Code, Section R 14-5-202(R).
1149

We request that a minimum component size for leak survey should be defined (i.e., line size).
We suggests [sic] that components associated with line size or tubing less than 2-inch diameter
should be excluded consistent with other regulations for leak detection and repair programs.
Other alternatives in addition to optical imaging cameras should be allowed for leak detection.
Limiting leak detection to the optical camera alone is too restrictive and would result in
increased costs and it would immediately create a shortage of qualified personnel due to the large
number of entities required to perform this leak monitoring. Accepted alternatives such as
methane detectors, ultrasonic / acoustical methods, and soap solution are available, are already
used in practice, and provide better or equivalent data in many cases. These methods should be
included in 98.234(a), and where standard methods are not readily available, industry standards
or best practices should be allowed to accommodate those entities that cannot obtain the required
equipment or manpower resources within the implementation timeline of this rule.
response to comment EPA-HQ-OAR-2009-0923-1039-18. Todays final rule specifies that
tubing systems equal to or less than one half inch in diameter neednt be monitored for
equipment leaks or reported under subpart W. Please see the response to comment EPA-HQ
OAR-2009-0923-1152-8.

Section 98.233(q)(2) Leak Detection and Leaker Emission Factors (Onshore Natural Gas
Processing Facilities. API requests that a minimum component size for leak survey should be
defined. API suggests that components associated with line size or tubing less than 2-inch
diameter should be excluded because these components are expected to have minimal leakage
while significantly adding to the burden on industry for collection of data associated with this
smallest category of components. The leak detection and component count requirements should
also be consistent with existing leak detection and repair (LDAR) regulations for onshore natural
gas processing facilities to minimize additional burden required to collect data on these
components. In addition, other alternatives to optical imaging cameras should be allowed for
leak detection. Limiting leak detection to the optical camera alone is too restrictive and would
result in increased costs, as it would immediately create a shortage of qualified personnel due to
the large number of entities required to perform this leak monitoring. Accepted alternatives such
as methane detectors, ultrasonic / acoustical methods, and soap solution are available, are already
1150
used in practice, and provide better or equivalent data in many cases. These methods should be
included in Section 98.234(a), and where standard methods are not readily available, industry
standards or best practices should be allowed to accommodate those entities that cannot obtain
the required equipment or manpower resources within the implementation timeline of this rule.
0923-1039-18. Todays final rule specifies that tubing systems equal to or less than one half
inch in diameter neednt be monitored for equipment leaks or reported under subpart W. Please
see the response to comment EPA-HQ-OAR-2009-0923-1152-8.

Vent Measurement Subpart W Should Allow Continuous Direct Measurement of Manifolded
Vent Lines as an Alternative that Operators Can Elect to Employ
The previous sections discuss emission estimation methods for vented sources and INGAA
comments on these vented emissions. In addition to those source-specific comments, the
Proposed Rule should allow continuous direct vent flowrate measurement as an approved
alternative, with that option selected at the discretion of the owner/operator. In general, INGAA
recommends that direct measurement alternatives should not be precluded from Subpart W,
and flexibility should be provided to apply demonstrated methods with similar or higher
accuracy. Specifically, INGAA recommends that continuous direct measurement of vented lines,
including manifolded lines, should be allowed as an operator option under Subpart W. In this
case, the annual vented emissions would be measured and reported, and the report would identify
the sources addressed by the vent measurement. For a transmission compressor station, this
could include more than one of the seven sources (or more than one reciprocating compressor
operating mode).
Some facilities are equipped with existing manifolded vent lines, but this type of configuration is
not prevalent in the natural gas transmission and storage segments. At other facilities, vent access
issues discussed above may increase interest in developing alternative vent line configurations.
In the long term, issues with conducting annual measurements may result in some facilities
migrating to continuous direct measurement. A direct measurement approach for manifolded
lines would provide accurate annual data rather than projection of a one-time snapshot to annual
emissions. However, the Proposed Rule precludes this option. INGAA recommends rule
revisions to allow continuous direct measurement as an optional approach that the
owner/operator may elect to use. This alternative is consistent with data quality and inventory
accuracy objectives. In addition, the alternative will provide near-term options while providing
the flexibility to adapt and implement new technology.
For manifolded vent lines, the vent measurement would co-mingle streams from multiple sources
1151
(e.g., multiple compressors), such that mode-based or source-specific emissions (e.g.,

compressor vent versus compressor unit blowdown) would not be available. However, the
tradeoff for lack of this particular information for a facility is a more accurate annual report of
cumulative facility vented gas emissions. It appears that mode-based and source-specific
information is of interest to help inform policy objectives, and EPA may be concerned by
aggregation resulting from manifolded lines. However, the vast majority of existing facilities will
not employ continuous vent measurement, so source-specific data will be broadly available from
those facilities. Complementing such information with total vent volume measurements at other
facilities should actually provide EPA with more compelling and complete information on
facility GHG emissions to help inform future policy decisions.
In addition, measurement of manifolded vents is analogous to EPA provisions for general
stationary combustion sources that allow aggregate reporting of common flows i.e.,
98.36(c)(2) and (3) allow combustion sources to use aggregated volumes from common stacks
and common fuel lines.
INGAA recommends that EPA add another paragraph (i.e., subsection) to 98.233 to indicate
that direct, continuous measurement of vent lines, including manifolded lines, is an acceptable
alternative to the source-specific procedures identified in 98.233.
Requirements for this alternative measurement approach should include the following:
Owners and operators can elect to use continuous direct vent measurement as an alternative to
other estimation methods defined in 98.233.
Meters should be accurate to within 5%, as defined in Subpart A, 98.3(i).
As with vent measurement under 98.233(p), the use of any acceptable measurement device
should be allowed. For example, 98.233(p)(2)(ii) indicates, Use a temporary meter such as, but
not limited to, a vane anemometer or a permanent meter such as, but not limited to, an orifice
meter. [emphasis added] Any measurement device should be allowed that meets the accuracy
criteria in Subpart A, 98.3(i).
For manifolded vents that measure multiple sources, the company GHG Monitoring Plan would
identify the sources emitting gas through the common vent and ensure that all applicable sources
are addressed. The aggregated sources would also be identified in the annual report.
To address aggressive schedule implementation, operators should be allowed to start
continuous monitoring by October 1 of the first year i.e., 3 months of data would be used to
estimate annual emissions for the first reporting year.
INGAA recommends the following addition as 98.233(aa) to add this option to the Rule:
(aa) Alternative Continuous Monitoring of Common Vent Lines. In lieu of the methods defined
in 98.233, the owner or operator can elect to continuously monitor a vent line or lines that
includes emissions from one or more sources identified in 98.232 for the industry segment,
including common or manifolded vents lines that combine more than one source. The following
requirements apply.
(i) The GHG Monitoring Plan and annual report shall identify the 98.232 sources addressed.
The GHG Monitoring Plan shall describe the venting system, measurement basis, and data
acquisition system.
1152
(ii) The meter shall meet the requirements of 98.3(i).

(iii) Annual emissions shall be reported based on continuous monitoring throughout the year or
upon facility startup with the exception of the initial reporting year, when continuous monitoring
must be initiated no later than October 1 with annual emissions estimated based on three months
of continuous monitoring data.
Response: EPA disagrees with the comment on compressor manifolded vent measurement.
Please see response to comment EPA-HQ-OAR-2009-0923-1041-3. EPA has revised todays
final rule to allow permanent meters on compressor vents, but not manifolded vents. Subpart A,
and its instrument calibration procedures, applies to subpart W unless otherwise specified in
subpart W. Please see The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part
98) preamble Section L. EPA has revised todays final rule Section 98.234(b) to clarify this.

Organization: CMC Solutions and CMC Support
Commenter: Brian Swanson
Where continuous emission monitoring systems (CEMS) are required or may be required in the
future we would offer that as an alternative predictive emission monitoring systems (PEMS) can
be used as an alternative monitoring system. We are requesting that PEMS be written into the
rule as an alternative wherever CEMS are required.
PEMS are a very cost effective and robust alternative to a CEMS and has proven itself in many
40 CFR Part 60 and 40 CFR Part 75 applications
Response: Subpart W does not require the use of CEMS, so this comment regarding PEMS does
not apply.
Organization: CMC Solutions and CMC Support
Commenter: Brian Swanson
PEMS as an Alternative to CEMS
U.S. regulations require continuous emissions monitoring systems (CEMS) and allow for the use
of predictive approaches as an alternative providing the installed predictive emissions monitoring
system (PEMS) meets rigorous performance specification criteria and the site performs ongoing
quality assurance tasks such as periodic audits with portable analyzers and annual accuracy
testing. PEMS have been used in the U.S. for gas turbine compliance monitoring under 40 CFR
Part 60 for more than 20 years.
Prior to the promulgation of CEMS requirements in the U.S., process emissions were verified
using simple parametric equations or manual stack test procedures (U.S. EPA Reference
1153
Methods). Modern empirical PEMS have also evolved to meet the need for continuous
monitoring of process emissions at the lowest possible cost. There are performance specification
tests and periodic audits such as the annual relative accuracy test audit that are equivalent in cost
as those required of compliance CEMS, however, PEMS can provide years of service with little
or no ongoing operational or maintenance cost with a robust model developed and in place.
PEMS used for compliance are certified according to the rigorous standards specified using PS
16 (Part 60) or Subpart E (Part 75). These certifications require the PEMS to pass a relative
accuracy test audit (RATA). In addition to these tests, SmartCEM provides additional quality
assurance through EPA mandated performance tests. These include online input validation, drift
analysis, input failure analysis, statistical analysis, model operating envelope analysis, and other
automated procedures and tests as specified in 40 CFR Part 75, Subpart E and in 40 CFR Part 60,
Appendix B, PS-16. The test results are used to evaluate a SmartCEM PEMS model prior to
deployment and are presented to regulatory agents as part of the certification process. All
performance tests are completed prior to RATA testing, such that the reference method RATA
test witnessed by the regulatory agency is the final step in the procedure to achieve certification
of a 40 CFR Part 60 compliance PEMS or for a 40 CFR Part 75 Subpart E application with the
exception of the required report submittal and petition to U.S. EPA Administrator
Attachment C. Safety Issues Associated with GHG Reporting Rule Data Collection for Onshore
Petroleum and Natural Gas Production
Test methods/procedures and job hazards assessments need to be completed to evaluate safe
monitoring and measurement from Subpart W sources. EPA has not adequately contemplated the
costs or time required to implement source measurements further discussed below. Noble
requests that any source deemed unsafe to measure be placed on a list and accompanied with an
explanation of the specific safety concerns. In the absence of measured data, Noble suggests
using published emission factors or engineering estimates to be used in lieu of unsafe to monitor
sources. Alternatively EPA should provide a method or procedure that allows the safe access and
measurement of such sources.
Reciprocating Compressor Rod Packing Vents
Safety is a very significant issue when attempting to collect measurements from roofline vents at
onshore natural gas processing facilities or gathering compressor stations. Safe access to and
sample collection from leaking reciprocating compressor rod packing vents are mandatory. At
these existing facilities, the majority of vents are routed outside the compressor building and
elevated above the roof line to disperse potentially flammable gas vapor. In addition to gaining
safe access to the elevated vent source, these vents are frequently manifolded with other vents or
1154
adjacent to blowdown vents that may automatically discharge to relieve pressure. Any
measurement requirement that results in placing test personnel in a potentially dangerous
environment is deemed unacceptable.
Resolution would likely require modifications to provide vent access from a safer location.
However, line accessibility may not be readily available as the facilities were not constructed
considering access. Engineering evaluations and analysis would be necessary to identify possible
sample port locations. Due to the proposed schedule, this would put operators at risk of failing to
comply with the annual survey requirement, or placing test personnel would be exposed to an
immediate danger defying safety procedures. Therefore, additional time is required to implement
the vent measurement program so that modifications can be made to accommodate measurement
from a safer location than the vent exhaust from these elevated sources.
The majority of building roofline vents would be deemed inaccessible under conventional LDAR
programs, as such components would be considered unsafe-to-monitor (UTM) or difficult-to
monitor (DTM) in a typical LDAR program (i.e., components that cannot be monitored without
elevating the monitoring personnel more than 2 meters above a support surface). In addition, a
manlift is typically required to gain access over high pressure yard piping further complicating
access and safety concerns for elevated vents. Under a typical LDAR program, monitoring can
be deferred due to unsafe or difficult-to-monitor components. In response, the facility must
maintain documentation that explains the conditions under which the components become safe to
monitor or no longer difficult to monitor. For vent lines at most gathering compressor stations,
typical conditions under which these vents would be safe would preclude normal engine
operations or pressurized scenarios where the potential for a vented release is possible.
Leak detection and repair (LDAR) programs are required as part of 40 Code of Federal
Regulations (CFR) Part 60 (NSPS), 40 CFR 61 (NESHAP), 40 CFR 63 (MACT), and 40 CFR
264 (Hazardous Waste Handling). LDAR requirements provide allowances for identification and
explanation for any equipment that is:
- unsafe to monitor (UTM) Exposing test personnel to an immediate danger as a consequence
of monitoring. Unsafe access includes, but is not limited to, the use of a wheeled scissor-lift on
unstable or uneven terrain, the use of a motorized man-lift basket in areas where an ignition
potential exists, or access would require near proximity to hazards such as electrical lines, high
pressure piping, proximity to heated, sharp, or rotating equipment, or would risk damage to
equipment.
- difficult to monitor (DTM) obstructed access (e.g. insulated), elevated access > 2 meters
required, limited access to component
- inaccessible - Obstructed by equipment, piping, or insulation that prevents access to the
connector by a monitor probe; Buried or confined space entry prohibits access to measurements.
DTM and UTM provisions for accommodating safety related concerns or measurement or
monitoring access need be considered and allowances for estimating these sources provided.
1155
Pressurized Vessel Liquid Sample Collection

Care must be taken to safely obtain liquid samples from pressurized vessels. These pressure
vessels have a wide range of operating temperatures and pressures. When mishandled, samples
taken from high-pressure systems can cause serious injury, or even death. Even a small pinhole
in a high-pressure line can cause serious injury. Whenever a sample is drawn, every precaution
necessary must be taken to ensure the safety not only of the sample collector, but also of those
working around the system. As with all high-pressure sampling systems, appropriate safety
precautions must be followed. Sampling procedures and taps must be properly sighted and
installed to reduce safety concerns. The cost and burden for safely accessing these samples has
not been considered or addressed.
Other Related Safety Concerns
Safety training for fall protection and hydrogen sulfide (H2S) or respirator certification will be
required for most test personnel to allow safe measurement from vents and component
population counts. Unlike the downstream sector, field natural gas may contain higher levels of
hydrogen sulfide (highly toxic and flammable gas) and higher quantities of nitrogen or carbon
dioxide (potential asphyxiants). Potential exposure to these high concentrations of these
compounds requires training, safety awareness, proper certifications, and field/facility overviews
prior to conducting any measurement activities. In addition, hot work permits are required for
any equipment or instrument that is not intrinsically safe. This permit adds to the cost and
implementation burdens and has not been considered by EPA in the cost analysis.
Response: EPA recognizes the importance of ensuring safety, and has added alternative
methodologies to ensure safety. Please see response to comment EPA-HQ-OAR-2009-0923
1024-11. For certain sources EPA will allow the application for the use of best available

Commenter:
75 FR 18621: 1. Direct Measurement
Comment: WBIH strongly requests EPA to consider flexibility with monitoring and
measurement methods of vents to ensure the safety of personnel conducting these tasks.
WBIH's number one priority is the safety of our employees and contractors providing service at
our facilities. This proposed rule requires very prescriptive requirements for monitoring and
measurement of components requiring access to vents located outside, in high locations, both on
tanks and building roofs. Each WBIH comment has been developed with safety at the forefront
of the process
1156
1024-11.
Safety Concerns
The proposed rule does not allow exemption's [sic] for difficult or unsafe to monitor
components. Under the previously proposed Subpart W "component fugitive emissions sources
that are not safely accessible within the operator's arm's reach from the ground or stationary
platforms are excluded." EPA had acknowledged that some of the components included in the
inventory would be located at areas that are either unsafe to access or located such that gaining
accessibility would require additional and costly measures such as cranes or lifts. The unsafe
difficult to monitor provisions should be incorporated into the currently proposed Subpart W to
ensure the safety of personnel and limit the undue burden of reporting
1024-11.

Commenter: Kim Dang
EPA should not require reporting from individual components that cannot be monitored without
placing employees safety at risk, or that are physically inaccessible.
1024-11.
Commenter:

Direct measurement is not appropriate for any sources.
1157
GPA feels that direct measurement is not appropriate for any sources for the purposes of GHG
emission inventories, consistent with other emission inventory practices. Although EPA has
reduced direct measurement requirements in the current proposal for Subpart W, EPA continues
to employ methods that ignore longstanding and accepted international protocols for emission
inventory calculations. The API Compendium in particular has been accepted as an international
standard for estimating GHG emissions across the oil and gas industry. For example, Australia
NGERS, EU-ETS, and Alberta emission inventories are all based on the API Compendium. EPA
is assuming that a significant increase in overall accuracy will result by requiring more complex
and costly estimation of GHG emissions from even the smallest sources within the natural gas
gathering and processing industry. This assumption is made while overlooking the EPAs own
data that indicates the natural gas gathering and processing industry emits only a very small
fraction of the total US GHG emissions annually. In addition, the currently proposed Subpart W
ignores the standards of statistical averaging.
While GPA supports the full use of the API Compendium to calculate GHG emissions for all
sources covered by the proposed Subpart W, at a minimum we suggest the application of the API
Compendium in place of any direct measurement methods. For the gas gathering and processing
sector, direct measurement is generally proposed to be limited to compressor wet seal vents and
rod packing venting. The current proposal potentially requires direct measurement of compressor
rod packing vents from every cylinder of a compressor in cases where the vent lines are not tied
to a common line, which is a common practice in gathering compression facilities and
production facilities. The prescribed methods for these sources is flawed by assuming a snapshot
measurement applied across an entire year is more accurate than an emission factor based on
well documented and controlled studies.
Response: EPA disagrees with the comment that direct measurement is not appropriate for any
source. Please see The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98)
Section II.L, and the Technical Support Document (TSD) for todays final rule found in docket
(EPA-HQ-OAR-2009-0923). EPA disagrees that the API Compendium is suitable for all sources
covered by subpart W. Please see response to comment EPA-HQ-OAR-2009-0923-1206-47.
Organization: Texas Pipeline Association
Commenter: Patrick J. Nugent
Direct measurement requirements should be eliminated from the rule. TPA
objects to any requirement for direct measurement of emissions from equipment covered by
Subpart W. Direct measurement of fugitives is the most cost-intensive component of the
proposed rules. In addition, EPA should consider the fact that there would be significant
greenhouse gas emissions associated with the burdensome efforts required to implement and
perform direct measurement protocols, e.g. emissions caused by sending crews all across the
State of Texas to perform the direct measurement required by the rule. Further, direct
measurement of emissions only shows the emissions levels that exist at a single point in time,
meaning that data collected by direct measurement methods is not representative of ongoing
annual emission from the sources at issue.
1158
Simpler and much more efficient data collection may be obtained through the use of
methodologies such as those presented in the Compendium of Greenhouse Gas emissions
Estimation Methodologies for the Oil and Gas Industry, developed by the American Petroleum
Institute, dated 2004, or as subsequently revised ("API Compendium"). TPA strongly urges EPA
to revise Subpart W to allow for the estimation and calculation of fugitive emissions through the
use of approved factors in place of the onerous detection and measurement provisions that are
being proposed. Methodologies such as are provided in the API Compendium are used in
connection with permits and emissions inventories, and as such they should be sufficient for use
in connection with a reporting rule such as Subpart W.
Response: Please see response to comment EMAIL-0002-9 (comment also located in
rulemaking memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923).
Pre-amble, page 21, "In this supplemental proposal, EPA is requiring the use of direct
measurement of emissions for only the most significant emissions sources where other options
are not available, and proposing the use of engineering estimates, emissions modeling software,
and leak detection and publicly available emission factors for most other vented and fugitive
sources."
- Given the sheer number of sources for which emission estimates are required, CAPP supports
the use of engineering estimates, emission modeling software and publicly available emission
factors as an alternative to direct emission measurement. Direct measurement of GHG emissions
is technically challenging and impractical for the scale of reporting that is envisioned due to the
basin level reporting approach. In addition, the direct measurement of some emission sources
results in safety issues. While the proposed emission estimation methods represent a much more
feasible approach, the required volume of data gathering, sampling and analysis is still daunting.
Response: EPA disagrees with the comment that direct measurement is not appropriate. Please
see The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98); preamble
(EPA-HQ-OAR-2009-0923). EPA disagrees with the comment regarding onshore production
cost. Please see response to comment EPA-HQ-OAR-2009-0923-1011-29 and the preamble
Section III.B. EPA recognizes the importance of ensuring safety, and has added alternative
1024-11.
1159

There Are Other Means to Get Meaningful Data
We agree with the suggestion posed by oil and gas sector organizations: there are other
emissions estimating tools available to assist EPA in gauging the impact of oil and gas-related
emissions. EPAs own Natural Gas Star program is one such tool. Industry is also developing
methodologies and tools to calculate GHG emissions for the sector. 148 These approaches seem a
better path to pursue than expansive regulatory requirements that will fall disproportionately on
small businesses.
Response: EPA disagrees that the API Compendium is appropriate for all sources; please see
the response to comment EMAIL-0002-9 (comment also located in rulemaking memo Early
Comment Submissions in docket EPA-HQ-OAR-2009-0923). EPAs Natural Gas STAR
Program, while valuable, is a voluntary tool and would not provide the comprehensive data
across industry segments and encompassing emissions sources as would be received through the
GHG reporting program. Also, Gas STAR data is not based on set data collection and
measurement methodologies and the data reported by partners are emissions reduced, not total
emissions as will be collected under the MRR. Please see The Final Mandatory GHG Reporting
Rule (Final MRR), (40 CFR part 98) preamble Section I.D.
The Rule Should Include Alternative Tools for Vent Measurement and Vent Emissions Detection
To provide flexibility, the Final Rule should allow all reasonable technologies and methods for
measuring leaks from reciprocating compressor rod-packing vent using a suite of technologies
and methods to provide flexibility. Alternative tools (e.g., high volume sampler) for vent
measurement should also be included provided that the vent rate is within the acceptable range of
the instrument. Thus, 98.233(p)(2) should reference 98.234(c) to allow use of calibrated bags
and 98.234(d) to allow use of a high volume sampler. The Proposed Rule should also reference
98.234(a)(1) within 98.233(p)(2)(i) to allow optical camera screening of elevated vents prior
to measurement.
INGAA requests that EPA broaden the list of allowed detection and measurement tools to
include all demonstrated technologies and alternatives that are used in practice. EPA studies have
employed all of the recommended measurement and leak detection technologies discussed below
in previous field studies. These tools provide the necessary flexibility to screen leaking
components via alternative techniques that may be more suitable to ambient conditions, available
equipment, or survey team experience level.
148
See, for example, the American Petroleum Institutes Gas Emissions Estimation Methodologies for the Oil and
Gas Industry.
1160
Since EPA proposes to require optical gas imaging for fugitive component screening and natural
gas transmission condensate tank vent monitoring, INGAA is curious why EPA elected to
exclude this technology for detecting leaks from elevated vent sources. Screening the compressor
rod-packing vent using an optical gas imaging instrument would provide a safer method for
detecting leakage through roofline vents and eliminate the need to measure non-leaking,
difficult-to-access vents. Unsafe or difficult-to-monitor compressor rod-packing vents with
visible leaks could be measured using conventional sampling or metering if the vent can be
accessed at an alternative location in the line.
Safe access to measure roofline vent lines and condensate tank vents, or alternative sampling
locations, will need to be integrated into existing facilities. Thus, during the initial year, INGAA
recommends phasing in these measurements (see Comment XII). INGAA recommends reporting
combustion and some vented emissions starting in year one and phasing in measurement and
monitoring over three years. Other phase-in options include precluding inaccessible vents from
reporting during initial years, or allowing vent screening using optical imaging. In the latter case,
if the camera detects vented emissions, a measurement would be completed if the vent is safely
accessible. If not accessible, vent status would be noted in the report and measurement would be
initiated in the second or third year after safe access has been established.
EPA should be familiar with these alternative tools and methods as all been have been used to
identify, quantify, and/or report venting losses under other EPA programs including Natural Gas
STAR.
0923-1039-18. EPA agrees that the reciprocating compressor rod packing monitoring method
Section 98.233(p)(2)(i) should include references to Section 98.234 for both calibrating bagging
and high volume sampler methodologies, and todays final rule has been revised. However, EPA
disagrees that it should require optical gas imaging leak detection in Section 98.233(p), as all
vents must be measured. EPA recognizes the importance of ensuring safety, and has added
alternative methodologies to ensure safety. Please see response to comment EPA-HQ-OAR
2009-0923-1024-11. EPA disagrees with the comment regarding a phased-in approach.
However, EPA has determined that for specified emissions sources for certain industry segments,
some reporters may need more time to comply with the monitoring and QA/QC requirements of
Subpart W by January 1, 2011. For certain sources EPA will allow the application for the use of
best available monitoring methods. Please see the preamble Section II.F.
The reporting burden can be significantly reduced without compromising data quality by: (1)
establishing de minimis cutoffs for certain operations; (2) allowing emissions estimates based on
the API Compendium for upstream production sites with emissions of less than 3,000 metric tons
1161
per year ("tpy") CO2e; (3) providing more flexibility in determining emissions for upstream
production sites with emissions of more than 3,000 tpy CO2e; (4) eliminating reporting of
emissions from portable non-self-propelled equipment because EPA can develop a reasonable
estimate based simply on the number and type of wells completed; and (5) eliminating direct
measurements, which would be inordinately costly and would not provide better data quality
than existing, widely accepted estimating techniques.
Response: Regarding the comments on de minimis and revised methodologies in todays final
rule, please see response to comment EPA-HQ-OAR-2009-0923-1018-2. EPA disagrees with
the comment on a threshold for onshore production sites. Please see the Technical Support
Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923). EPA
disagrees with the comment that direct measurement is not appropriate. Please see The Final
Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.L, and
the Technical Support Document (TSD) for todays final rule found in docket (EPA-HQ-OAR
2009-0923). EPA disagrees with the comment regarding onshore production cost. Please see
response to comment EPA-HQ-OAR-2009-0923-1011-29 and the preamble Section III.B. EPA
recognizes the importance of ensuring safety, and has added alternative methodologies to ensure
safety. Please see response to comment EPA-HQ-OAR-2009-0923-1024-11. EPA disagrees
with the comment on combustion emissions from portable equipment. Please see the Technical
Support Document EPA-HQ-OAR-2009-0923.

Additionally, the API Compendium in particular has been accepted as an international standard
for estimating GHG emissions across the oil and gas industry. For example, Australia NGERS,
EU-ETS, and Alberta emission inventories are all based on the API Compendium. EPA is
assuming that a significant increase in overall accuracy will result by requiring more complex
and costly estimation of GHG emissions from even the smallest sources within the natural gas
gathering and processing industry. This assumption is made while overlooking the EPAs own
data that indicates the natural gas gathering and processing industry emits only a very small
fraction of the total US GHG emissions annually. In addition, the currently proposed Subpart W
ignores the standards of statistical averaging.
While GPA supports the full use of the API Compendium to calculate GHG emissions for all
sources covered by the proposed Subpart W, at a minimum we suggest the application of the API
Compendium in place of any direct measurement methods. For the gas gathering and processing
sector, direct measurement is generally proposed to be limited to compressor wet seal vents and
rod packing venting. The current proposal potentially requires direct measurement of compressor
rod packing vents from every cylinder of a compressor in cases where the vent lines are not tied
to a common line, which is a common practice in gathering compression facilities and
production facilities. The prescribed methods for these sources is flawed by assuming a snapshot
1162
measurement applied across an entire year is more accurate than an emission factor based on
well documented and controlled studies.
Response: EPA disagrees with the comment that direct measurement is not appropriate. Please
see The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble
(EPA-HQ-OAR-2009-0923). EPA disagrees that the API Compendium is suitable for all
sources covered by subpart W. Please see response to comment EPA-HQ-OAR-2009-0923
1206-47. As explained in Section II.E of the preamble, EPA has decided not to include
monitoring and reporting requirements for gathering lines and boosting stations at this time.

Direct Measurement
We support the use of direct measurement for all source categories as our preferred measurement
and reporting method, because it is the most accurate method. The petroleum and natural gas
industry has some of the most economically attractive alternatives to control emissions at a
significant profit, and is a source category where emissions are, by EPAs own admissions,
dramatically underestimated. For this reason, we believe that more accurate emission
measurement should be required. Requiring more accurate measurement will provide further
economic incentive for Operators to install profitable emission control.
EPA has modified the Subpart W rule from its original proposal 35 to rely much less on direct
measurement and more on engineering estimates developed by computer models, emission
factors and engineering calculations. 149
For all of the same reasons identified in our comments on EPAs original MRR proposal, 150 we
strongly support the use of direct measurement wherever possible, with a high priority on its use
for complex systems and those with a high potential for fugitive emissions, such as are found in
the oil and gas sector. EPA must, as previously stated in our comments on the original rule,
develop mechanisms to move the reporting system steadily towards the most accurate and
precise reporting methods possible which will mean moving towards the use of more direct
measurement methods and will also mean systematically and periodically auditing and reviewing
the proposed reporting requirements to ensure continuous improvement in the reported data.
The difference between emissions measured directly and those that are estimated through
engineering methods can be significant. Unless direct measurement techniques are employed,
unintended emissions will go undetected and the emissions estimates, in general, cannot be as
149
See 75 Fed. Reg. 18625, April 12, 2010.
150
EPA-HQ-OAR-2008-0508-0635.
1163
accurate and precise as those obtained through actual measurements. In fact, this sector, in
particular, has seen, and EPA has documented in its Technical Support Document (TSD) for this
proposed rule, an enormous uncertainty in emissions estimates over the years. The degree to
which weve seen emissions estimates grow suggests there is a clear need to continually monitor
the quality of the data being reported. This need comes from the high level of uncertainty in
emissions characterization for this particular source sector. In the absence of requirements for
direct measurement, EPA should take every possible action to ensure that emissions factors and
estimates used are as accurate as possible by developing a sampling and audit program, as
discussed at #10 below.
Response: Please see response to comment EPA-HQ-OAR-2009-0923-1155-12. EPA disagrees
with the comments on auditing, please see EPAs response on verification in The Final
Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.N.

Emissions Are Seriously Under Estimated for this Source Category
A primary concern for the petroleum and natural gas systems is that EPA and other scientists
have documented a serious trend of emissions under-reporting for this source category.
Emissions underreporting is well documented in EPAs own Technical Support Document
(TSD) for the proposed MRR, where EPA concludes that several emissions sources are
significantly underestimated in the 1996 U.S. GHG Inventory. 151 More specifically, EPA
identified concerns with emission accuracy for the following sources: (1) well venting for liquids
unloading; (2) gas well venting during well completions; (3) gas well venting during well
workovers; (4) crude oil and condensate storage tanks; (5) centrifugal compressor wet seal
degassing venting; and (6) flaring. According to EPA, the emissions estimates for these sources
do not correctly reflect the operational practices of today and, in fact, EPA believes that
emissions from some sources may be much higher than currently reported in the U.S. GHG
Inventory. 152
In the TSD, EPA includes revised emission factors for four of these underestimated sources
leading to revised emissions estimates ranging from ten times higher (for well venting from
liquids unloading) to 35 times higher (from gas well venting from conventional well
completions) to as much as 3,500 and 8,800 times higher (for gas well venting from completions
and well workovers of unconventional wells, respectively).153 Overall, the revisions to just these
three sources and compressor wet seals would result in a more than 100% increase in estimated
151
These errors owe to errors in the underlying source data which in turn comes from the 1996 EPA/GRI/Radian
Greenhouse Gas Emission Inventory. See EPA Technical Support Document at p.7.
152
See EPA Technical Support Document at p.23.
153
See EPA Technical Support Document, Table 1, at p.8 and Appendix B.

1164
emissions from the oil and gas production sector and a 67% increase in estimated emissions from
the transmission and storage sector. 154
These enormous underestimates are, alone, reason enough to insist on more reliance on direct
measurement of emissions from these sources and systematic auditing of reporting methods.
However, there is even more evidence that the emissions estimates for these and other sources
remain uncertain and continue to be greatly underestimated.
An area of particular uncertainty is gas well completions and well workovers. As EPA notes in
the preamble to the proposed reporting rule:
[N]o body of data has been identified that can be summarized into generally applicable
emissions factors to characterize emissions from these sources [(i.e., from well completion
venting and well workover venting)] in each unique field. In fact, the emissions factor being used
in the 2008 U.S. GHG Inventory is believed to significantly underestimate emissions based on
industry experience as included in the EPA Natural Gas STAR Program publicly available
information (http://www.epa.gov/gasstar/). In addition, the 2008 U.S. GHG Inventory emissions
factor was developed prior to the boom in unconventional well drilling (1992) and in the absence
of any field data and does not capture the diversity of well completion and workover operations
or the variance in emissions that can be expected from different hydrocarbon reservoirs in the
country. 155
The 2008 U.S. GHG Inventory emphatically states that [n]atural gas well venting due to
unconventional well completions and workovers, as well as conventional gas well blowdowns to
unload liquids have already been identified as sources for which Natural Gas STAR reported
reductions are significantly larger than the estimated inventory emissions. 156 And, in response
to questions from Environmental Defense Fund, EPA responded that: presently, [reduced
emission completions (REC)] reductions reported in the Natural Gas STAR body of work is
larger than well completion venting in the inventory on an annual basis. 157 Specifically, the
U.S. GHG Inventory is based on an emission factor of a little over three thousand standard cubic
feet (3 Mcf) per gas well drilled and completed. 158 Yet, Natural Gas STAR program partner
experience shows several cases where emission factors were thousands of times higher than that
shown in the 2008 inventory. Examples include: (1) a BP project employing green completions
154
See EPA Technical Support Document, Table 2, at p.9.
155
75 FR 18621
156
Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006, p.
3-47.
157
See Environmental Defense Fund (EDF) Questions & USEPA Answers June 1, 2010 posted to the Docket
for this action on June 2, 1010, Docket ID EPA-HQ-OAR-2009-0923-0070.
158
Table A- 118: 2008 Data and CH4 Emissions (Mg) for the Natural Gas Production Stage, p. A-144, Inventory of
U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006. 11 See Natural
Gas STAR Program Recommended Technologies and Practices for Wells at
http://www.epa.gov/gasstar/tools/recommended.html, and specifically
http://www.epa.gov/gasstar/documents/green_c.pdf, slide.
1165
at 106 wells and reporting 3,300 Mcf of gas recovered per well; 159 All of these examples include
gas recovery estimates more than 1,000 times higher than the 3 Mcf of gas per well estimated in
the U.S. GHG Inventory for 2008. These data are consistent with the unconventional gas well
completion and workover data presented in EPAs TSD for the MRR. Specifically, Appendix B
on pp. 79-82 includes four examples from the Natural Gas STAR program with gas completion
rates of 6,000 Mcf, 10,000 Mcf, 700 Mcf and 20,000 Mcf per completion. EPA used an average
of these four data points for its emission factor for this source (i.e., 9,175 Mcf/completion).
Using this factor resulted in estimated emissions from completions of conventional and
unconventional wells of 86 billion standard cubic feet (Bcf). In comparison, the U.S. GHG
Inventory reports emissions of just 0.1 Bcf of gas from drilling and well completions. 160
More generally, Natural Gas STAR partners reported recovering between 7 and 12,500 Mcf
(average of 3,000 Mcf) of natural gas from each cleanup with the potential for an estimated 25
Bcf of natural gas recovery from green completions annually in 2005 compared with the annual
U.S. GHG Inventory total for Drilling and Well Completion of just 0.1 Bcf in 2008. 161 ,162 And
more recently, EPAs Natural Gas STAR program attributed 45 Bcf of gas to Reduced Emissions
Completions (RECs) in 2008 (representing 50 percent of EPAs Natural Gas STAR programs
annual total reductions).163 If, in fact, the emission factor for well completions were at least
1,000 times higher than what is reported in the U.S. GHG Inventory (e.g., if it were at least 3,000
Mcf instead of 3 Mcf) this would add 100 Bcf to the total estimated emissions from natural gas
systems, raising the total from 240 Bcf to 340 Bcf in 2008 (a 40% increase). 164
A New York Times article published on October 15, 2009, reports that EPA is currently
159
See attached 2009 workshop presentation by Devon, slides 3 and 13. 6,300 Mcf = 11.4 Bcf / 1,798 wells. ] (2) a
Devon Barnett Shale project employing green completions at 1,798 wells between 2005 and 2008 and reporting
6,300 Mcf of gas recovery per well;12 and (3) a Williams project employing green completions at 1,064 wells in
the Piceance Basin reporting 23,000 Mcf of gas recovered per well.[Footnote 13: See Natural Gas STAR Program
Recommended Technologies and Practices for Wells at http://www.epa.gov/gasstar/tools/recommended.html, and
specifically http://www.epa.gov/gasstar/documents/workshops/pennstate2009/robinson1.pdf, slide 16. We note
that the reported Williams reductions appear to assume an average of 32 days of uncontrolled venting during
flowback, which seems like an unreasonably long length of time.
160
See Table A-125: CH4 Emission Estimates from the Natural Gas Production Stage Excluding Reductions from
the Natural Gas STAR Program and NESHAP regulations (Gg), p. A-151, Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006. Drilling and Well Completion (2008) =
2.05 Gg. (2.05 Gg * 0.052 ft3/g = 0.1 Bcf).
161
See http://www.epa.gov/gasstar/documents/green_c.pdf, slides 9 and 5.
162
2.05 Gg.
163
See 2009 EPA Natural Gas STAR Program Accomplishments, available online at
http://www.epa.gov/gasstar/documents/ngstar_accomplishments_2009.pdf. Total sector reductions (2008) = 89.3
Bcf of which 50% are the result of RECs (50% of 89.3 Bcf = 45 Bcf).
164
See Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006,
Chapter 3, Table 3-38 CH4 Emissions from Natural Gas Systems (Gg), p. 3-45. Total emissions in Bcf: 4,591 Gg
* 0.052 ft3/g = 240 Bcf.
1166
reviewing and revising its estimate of methane emissions from U.S. gas wells. 165 According to
the article:
An E.P.A. review of methane emissions from gas wells in the United States strongly implies
that all of these figures may be too low. In its analysis, the E.P.A. concluded that the amount
emitted by routine operations at gas wells not including leaks like those seen near Franklin
is 12 times the agencys longtime estimate of nine billion cubic feet. In heat-trapping potential,
that new estimate equals the carbon dioxide emitted annually by eight million cars. 166 In fact,
the TSD for the MRR includes an estimate of 120 Bcf for U.S. completion and workover venting
(2007). As previously discussed, this estimate is based on an emission factor of 9,175 Mcf per
completion or workover. Given the sparse data (4 data points) and the enormously wide range of
potential emission factor values from this source (ranging from 700 to 20,000 Mcf) even the
recently revised estimates may very well continue to underestimate emissions from this source.
While it is clear that EPA acknowledges the lack of accurate emissions factors for well
completions, the enormous variation more than three orders of magnitude - between different
work sites and between work sites and EPAs suggested emissions factors demonstrates that
need for at least some direct measurement and study sufficient to develop more reliable
emissions factors.
The statistical representation of the U.S. GHG Inventory data includes a 95 percent certainty
range within which emissions from this source category are likely to fall for the year 2008. This
range, for Natural Gas Systems, includes a lower bound of -24% and an upper bound of
+43%. 167 As noted in the uncertainty analysis, [t]he heterogeneous nature of the natural gas
industry makes it difficult to sample facilities that are completely representative of the entire
industry. 168 And as a result, basing an emission factor on only a few representative sources
when considering the highly variable rates measured among these sources, results in a potentially
high degree of uncertainty as reflected in the reported uncertainty range.
Clearly, the EPA-acknowledged discrepancies between the inventory emissions reported to-date
and the Natural Gas STAR reported reductions and the high degree of uncertainty in both of
these data sources must be reconciled with rigorous reporting rule requirements for the oil and
gas sector and, in particular, for vented emissions from wells. And in order to place more
certainty on the numbers being reported EPA will need to employ a rigorous sampling and
auditing program to verify reported emissions. This particular source category demonstrates the
desperate need for more reliability in emissions reporting from the oil and gas sector.
Onshore production and processing storage tanks are an additional source with a potential for
high uncertainty in estimated emissions. A study prepared for the Texas Environmental Research
165
See www.nytimes.com/2009/10/15/business/energy-environment/15degrees.html#.
166
167
Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006,
Chapter 3, Table 3-41, p. 3-46.
168
Chapter 3, p. 3- 46.
1167
Consortium measured emissions rates from several oil and condensate tanks in Texas and
developed average emission factors based on direct measurement of vent gas flow rates. 169 The
study determined the direct measurement approach to be the most accurate for estimating oil
and condensate storage tank emissions at wellhead and gathering sites; however, other, less
accurate, approaches appear to be much more commonly used. EPA proposes the use of
modeling (E&P Tank) to calculate emissions from storage tanks in the MRR. Yet, the TSD
acknowledges significant weaknesses with this approach. 170 Again, only a move towards more
direct measurement (e.g., direct measurement of gas oil ratios, vent gas and flow rates) will
reduce these uncertainties in the reported data. Finally, as previously mentioned in our comments
on the original reporting rule proposal, Canadian natural gas processing plants discovered
methane emissions roughly an order of magnitude higher than estimated demonstrating the
critical need for comprehensive direct measurement of emissions from this significant sector of
GHG emissions. 171
Certain oil and gas sources are of particular importance due to the magnitude of their emissions.
These sources will require the best possible reporting requirements if the data gathered are to be
representative and of further use. In particular, uncertainties in data reported across a large
population of sources (e.g., the hundreds of thousands of pneumatic devices used throughout the
oil and gas sector) have the potential to greatly reduce the value of the dataset. Pneumatic
devices are a prime example of a source that is particularly important in terms of magnitude and
is also important due to the sheer number of sources and the associated compounding effects of
the uncertainty associated with each individual source. The magnitude and importance of this
emissions source support a greater emphasis on direct measurement (e.g., metering) and on
rigorous verification and audit for any reporting that will be based on engineering estimates (e.g.,
OEM emission factors).
Response: Please see response to comment EPA-HQ-OAR-2009-0923-1155-12. In addition,
EPA disagrees with the comment on verification. Please see The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.N.
169
Hendler A., Nunn J., Lundeen J., McKaskle R., VOC Emissions from Oil and Condensate Storage Tanks Final
Report, April 2, 2009, available online at
http://files.harc.edu/Projects/AirQuality/Projects/H051C/H051CFinalReport.pdf] The U.S. GHG Inventory
mentions this study but indicates that [b]ecause of the limited dataset and unexpected jumps in data points which
can be attributed to non-flashing emission affects, the United States decided that further investigation would be
necessary before updating the inventory emission factor.[Footnote 24: Inventory of U.S. Greenhouse Gas
Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006, Chapter 3, p. 3- 47.
170
See EPA Technical Support Document, at p.134.
171
See Allan K. Chambers et al., DIAL Measurements of Fugitive Emissions from Natural Gas Plants and the
Comparison with Emission Factor Estimates (also processing plants to be roughly an order of magnitude higher
than estimated emissions of volatile organic compounds and benzene).
1168

Natural gas processing plants Comments on the original reporting rule identified Canadian
natural gas processing plants that have discovered methane emissions roughly an order of
magnitude higher than estimated, demonstrating the critical need for comprehensive direct
measurement of emissions from this significant sector of GHG emissions 172
Response: EPA considered the monitoring method for each source and in several cases in
onshore natural gas processing it determined that direct measurement was not necessary. Please
see the Technical Support Document (TSD) for todays final rule found in docket (EPA-HQ
OAR-2009-0923).

EPA Should Place a High Priority on the Use of Direct Measurement and Should Implement a
Sampling and Audit Program to Provide for Periodic Improvements to the Proposed Data
Collection Program.
Under EPAs proposed reporting rule for the oil and gas sector, facilities would be required to
quantify greenhouse gas emissions according to a range of methods. Methods include the direct
measurement of emissions, engineering estimation methods (e.g., simulation models, engineering
calculations, original equipment manufacturer emission factors, and default emission factors),
leak detection and leaker emission factors as well as equipment count and population emission
factors.
In its supplemental reporting rule for the oil and gas sector, EPA has significantly modified its
original proposal. 173 We strongly support the use of direct measurement wherever possible,
especially for complex systems and those with a high potential for fugitive emissions (such as
are found in the oil and gas sector). EPA should develop mechanisms to move the reporting
system steadily towards the most accurate and precise reporting methods possible. This will
require moving towards the use of more direct measurement methods and systematically
reviewing the proposed reporting requirements to ensure continuous improvement in the reported
data.
The difference between emissions measured directly and those that are estimated through
engineering methods can be significant. Emissions estimates generally cannot be as accurate and
172
See Allan K. Chambers et al., DIAL Measurements of Fugitive Emissions from Natural Gas Plants and the
Comparison with Emission Factor Estimates (also finding fenceline measurements of Canadian natural gas
processing plants to be roughly an order of magnitude higher than estimated emissions of volatile organic
compounds and benzene), referenced in Comments EPA-HQ-OAR-2008-0508-0635.
173
See 74 FR 16448, April 10, 2009.] to rely much less on direct measurement and more on engineering estimates,
leaker factors and emissions factors [Footnote 18: See 75 FR 18625.
1169
precise as those obtained through actual measurements, and they allow for unintended emissions
to go undetected. The oil and gas sector has seen (and EPA has documented in its Technical
Support Document (TSD) for this proposed rule) an enormous uncertainty in emissions estimates
over the years. The degree to which emissions estimates have grow suggests there is a clear need
to continually monitor the quality of the data being reported.
Response: Please see response to comment EPA-HQ-OAR-2009-0923-1155-12. In addition,
EPA disagrees with the comment on verification. Please see The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.N.
The Number of Source Categories with Significantly Underestimated Emissions is Considerable.
The TSD for the proposed reporting rule includes a discussion of several emissions sources
believed to be significantly underestimated in the U.S. GHG Inventory. 174 Specifically, EPA
identifies the following sources: (1) well venting for liquids unloading; (2) gas well venting
during well completions; (3) gas well venting during well workovers; (4) crude oil and
condensate storage tanks; (5) centrifugal compressor wet seal degassing venting; and (6) flaring.
EPA states that the emissions estimates for these sources do not correctly reflect the operational
practices of today and that emissions from some sources may be much higher than currently
reported in the U.S. GHG Inventory. 175
In the TSD, EPA includes revised emission factors for four of these underestimated sources
leading to revised emissions estimates ranging from ten times higher (for well venting from
liquids unloading) to 35 times higher (from gas well venting from conventional well
completions) to as much as 3,500 and 8,800 times higher (for gas well venting from completions
and well workovers of unconventional wells, respectively) 176 , 177 Overall, the revisions to just
these three sources and compressor wet seals would result in a more than 100 percent increase in
estimated emissions from the oil and gas production sector and a 67 percent increase in estimated
emissions from the transmission and storage sector 178 .
These enormous underestimates are, alone, reason enough to require greater reliance on direct
measurement of emissions from these sources and systematic auditing of reporting methods. But
there is even more evidence that the emissions estimates for these and other sources remain
uncertain and continue to be greatly underestimated.
174
See Technical Support Document (TSD) at 7, citing EPA/GRI/Radian, 1996.
175
Id. at 23.
176
Id. at 23.
177
See Table 1 of TSD at 8 and Appendix B.
178
See Table 2 of TSD at 9.

1170

Organization: ENVIROTECH ENGINEERING
Commenter: Terence Trefiak
Direct Leak Measurement
It is our understanding that the revised proposed EPA rule has excluded the requirement to
quantify leak rates and instead have recommended using published emission factors base on
component type. We understand that this was done to try and reduce the perceived burden/cost of
completing a fugitive emission assessment. Based on our experience, we believe that direct
measurement of leaks is a crucial component to a successful fugitive emission managements
program and the costs of direct measurement are usually a small portion of the total cost of a
fugitive emission assessment.
Firstly, by measuring a leak we provide our clients with the economic information needed to
make appropriate decisions on repair activities. You can have two of the exact same components,
one leaking at high rate (5 cfm) and one leaking at a low rate (0.05 cfm). The priority and
urgency of repair will vary depending on the leak rate.
Response: EPA disagrees that direct measurement is applicable for all equipment leaks. Please
see the response to comment EPA-HQ-OAR-2009-0923-1155-12.
Organization:
Commenter: Anonymous
EPA should also require direct measurement of emissions. Using equations to estimate emissions
is inaccurate.
Response: EPA disagrees that direct measurement is appropriate for all emissions sources.
Please see the response to comment EPA-HQ-OAR-2009-0923-1155-12.

Organization:
Commenter: A. Bhaskar
EPA should require direct measurement of emissions. Using equations to estimate emissions is
inaccurate. We need direct measurements to ensure the data is accurate to develop the best
policy.
1171
Response: EPA disagrees that direct measurement is appropriate for all emissions sources.

Organization: Emerson Process
Commenter: Patrick Truesdale
Can coriolis meters which are installed/calibrated/maintained according to consensus industry
standards (API/AGA) be used to measure Natural gas and still gas flow for both Subpart C and
Subpart W facilities?
Response: This response to comment is only in regard to subpart W sources. Questions
regarding subpart C must be directed to EPA directly as that subpart is final. Under the proposed
subpart W, facilities would be able to use coriolis meters where such meter types are allowed
under subpart W, provided they meet the calibration and accuracy requirements in 40 CFR
98.3(i) and those set forth in subpart W Monitoring and QA/QC Requirements.
Organization: Leak Surveys Inc.
Commenter: David Furry, President and Owner
First, we strongly urge the agency to include some requirements for training and on-going
experience into the Subpart W reporting rule. Part 98.234(a)(1) stipulates Use an optical gas
imaging instrument for fugitive emissions detection in accordance with 40 CFR part 60, subpart
A, SECTION 60.18 (i)(1) and (2) Alternative work practice for monitoring equipment leaks.
Along with many in the industry, we thought that the lack of specific training and on-going
operation was an inherent problem potentially leading to unqualified technicians performing
surveys and, consequently, missing leaks. Our experience with our own training program has
shown that, until a technician has at least 250 eye hours behind the camera, the potential to
overlook leaks is relatively high.
In addition, the Texas Commission on Environmental Quality (TCEQ) also noted this deficiency
and has added a training provision to the Texas version of the AWP. Under the TCEQ AWP
proposed methodology 179 , it is recommended that:
(A) The operator of the optical gas imaging instrument must receive a minimum of 24 hours of
initial training on the specific make and model of the optical gas imaging instrument before
using the instrument for the purposes of this supplemental leak detection.
179
Section SECTION 101.153, Voluntary Supplemental Leak Detection Program in subsection 101.153(b)(4),
1172
(B) Operators using optical gas imaging instruments for this supplemental leak detection shall
comply with one of the following requirements for on-going training purposes:
(i) Operators shall attend an annual eight-hour refresher training class on the optical gas imaging
instrument used for this supplemental leak detection; or
(ii) Operators shall maintain a minimum of 100 hours per calendar year of hands-on operational
experience with the model of optical gas imaging instrument used for the supplemental leak
detection. Operators electing this option shall maintain a written log of the operator's operational
experience with the optical gas imaging instrument.
We would recommend that similar language be added to the EPA reporting rule, and we strongly
recommend that 40 hours of training in general optical imaging technology be stipulated instead
of 24 hours. However, we dont believe (as stipulated in the TCEQ proposed rule) that the
training be specific to the make and model of camera, but rather needs to be specific to basic
camera and leak technology. Additionally, we believe it is vital in obtaining quality surveys, that
any technician performing services under this rule must maintain at least 100 hours per year of
documented camera usage to maintain skills.
Response: EPA disagrees with adding mandatory training to the Alternative Work Practice for
40 CFR part 60, subpart A, Section 60.18. Todays final rule stipulates that reporters must
operate the optical gas imaging instrument in accordance with the instrument manufacturers
operating parameters and therefore EPA does not deem mandatory training necessary. Training
is not specified in subpart W. Subpart W does not supersede existing regulations and
requirements regarding training.
Section 98.233 (p) Reciprocating compressors rod packing, subsection (2) (i) Measure emissions
from all vents (including emissions manifolded to common vents) including rod packing unit
isolation valves and blowdown valves using bagging according to methods set forth in 98.234
(c). From our experience, the blow down vents seldom leak and we suggest that the optical
imaging camera be used from the ground to first determine which, if any of the vents need to be
measured. We also seek clarification on whether to use emission factors or direct measurement
on leaks found on a reciprocating compressor valves or gas found to be leaking around the
packing itself (vented to the atmosphere or to flare) but on the compressor.
Response: Todays final rule specifies that all compressor vents are to be measured. The rule
does not require the optical gas imaging camera to be used on vents, as this is unnecessary.
Compressor equipment leaks require leak detection and leaker emission factor. Please see
1173

Large Compressor B1owdown Valve Leak & Large Compressor Blowdown
Vent (Unit Isolation Valve Leak)
Subpart W proposes to require operators to report emissions from large compressor blowdown
valve leaks and large compressor blowdown vents (unit isolation valve leaks) using leak
detection with an optical gas imaging instrument. Optical gas imaging equipment is very
expensive (approximately $75,000-90,000 per camera) and may prove highly burdensome and
cost-prohibitive for many operators-and particularly smaller companies-to obtain. Such an
expense is particularly unreasonable if operators will be required to purchase such equipment
just to make the initial applicability determination (before it is determined that monitoring and
reporting under Subpart W is even required). While USEPA generally has assumed that
companies subject to the requirements of Subpart W will hire contractors to conduct much of the
monitoring and data collection required under the proposal, such an assumption may not be valid
in many cases (and, as noted above, IOGA-WV has concerns about the scant number of local and
regional contractors that currently have the capabilities and expertise necessary to perform the
analyses required by the proposed rule).
response to comment EPA-HQ-OAR-2009-0923-1039-18. Concerning the availability of
equipment or trained operators, for certain sources EPA will allow the application for the use of
best available monitoring methods. Please see the preamble Section II.F.
Furthermore, in the Technical Support Document (pp. 28-29), the EPA states, another key issue
is that all measurement technologies discussed require physical access to the emissions source in
order to quantify emissions. Many sources in a producing field (tanks, compressors, etc.) are
located in extremely remote areas and are therefore, by most definitions, inaccessible. Direct
sampling of those field sources will include previously unscheduled trips of environmental staff
which can take staff members off-site for days at a time, traveling in dangerous and remote areas
throughout the course of the year. Direct sampling of these sources is not only burdensome on
staff, it is dangerous for minimal benefit when emission factors could be developed for such
sources.
Companies will most likely rely on 3rd parties to conduct most direct measurement samples.
PAW requests guidance from the EPA on how to reasonably work with 3rd party vendors to
ensure their sampling methodologies meet EPA requirements.
1174
1024-11. EPA has revised several methodologies in todays final rule for onshore production to
reduce burden. Please see the preamble Section II.F. It is the responsibility of the reporter to
manage any third party vendors they may choose to contract for work. For further details on the
role of the designated representative please see the response to EPA-HQ-OAR-2009-0923-1024
16.
Organization:
Commenter: Michael James LeBrun
Recently we had a chance to look at a gas imaging camera made by FLIR. this [sic] model is the
GS320. I fell in love with this camera and have used it in a number of operations. Where my gas
sniffer increased a beeping tone to signify that gas was present, this camera lets you see a gas
plume as clear as day, and the amount could be less than a lighters amount of gas leaking, this
camera will find it. I have found a fair amount of leaks, small ones, but leaks on my equipment
of any kind is saving my operators money, and is puitting [sic] a smile on Mother Nature's
face!!! I can't stress enough that when you finalize this subpart w regulation, that somewhere in
BIG LETTERS, let our industry know that there is this wonderful new gas leak finding camera
made by FLIR, and a great service company called Western Thermal Imaging that has safe,
confident, well dressed, and well versed people on the equipment that we are using to produce
natural gas.
Response: EPA does not endorse technology vendors or service providers. Please see response
to comment EPA-HQ-OAR-2009-0923-1039-18.
98.233 (p):
In 98.233(p)(2)(ii) CAPP believes the use of an orifice meter is inappropriate in this application.
A better meter for this application would be an appropriately sized rotameter, diaphragm meter
or rotary meter (roots meter).
Response: The commenter has misinterpreted the rule. EPA does not require the use of an
orifice meter for measurement of reciprocating compressor vents. EPA has not prescribed the
type of meter that must be used. Please see response to comment EPA-HQ-OAR-2009-0923
1305-31.
1175

98.234 (c):
CAPP does not support the use of calibrated bags for all atmospheric emissions; the
methodology described by the EPA in 98.234(c) is very onerous and is not practical for sources
with many leak points or large surface areas such as compressor seal emissions.
Response: The commenter has misinterpreted the rule. The rule stipulates that calibrated bags
may be used only where the emissions are at near-atmospheric pressures such that it is safe to
handle and can capture all of the emissions, so if a leak cannot be totally captured and sealed,
then calibrated bags cannot be used for reporting under subpart W for that source.

Commenter: Kim Dang
Comments on the Proposed Monitoring and Quality Assurance / Quality Control Requirements.
As noted above, Kinder Morgan urges EPA to allow other forms of emissions detection apart
from optical imaging. The leak detection procedure in the proposed 40 C.F.R. SECTION
98.234(a) should be revised accordingly to accommodate appropriate standard leak detection
methods.
In addition, Kinder Morgan believes that the proposed 40 C.F.R. SECTION 98.234(a)(1) has a
grammatical error. The corrected provision should read Use an optical gas imaging instrument
for fugitive emissions detection in accordance with 40 C.F.R. part 60, subpart A, SECTION
60.18(i)(1) and (2), of the Alternative work practice for monitoring equipment leaks. Further,
EPA should clarify leak detection frequency by stating that, when used, optical gas imaging
would be performed on an annual basis.
0923-1039-18. EPA agrees that the proposed rule text be clarified. Please see the response to
EPA-HQ-OAR-2009-0923-1039-31. The commenter has misinterpreted the rule. The rule states
that annual leak detection surveys are required.
Unit -specific Emissions Estimation Methodologies
The unit-specific emissions estimation methodologies rely on a combination of direct
measurement and emissions factors. This approach must recognize that there are limitations of
1176
both methods. For emissions that are calculated based on emissions factors, most equations
include operating hours; most components have no way to determine operating hours and must,
therefore, assume 8,760 hours of operation per year, resulting in an artificially inflated inventory.
Where direct measurement is required, again, if a component is found to be leaking, the reporting
methods require that this emission rate is valid for the entire reporting period, when, in fact,
leaks are normally repaired as soon as possible. EPA should evaluate the possibility of
developing annual emission factors, rather than relying solely on hourly emission factors for
non-combustion units. This would allow emissions to be calculated based on the percentage of
time, i.e. days or months, that a facility operated annually rather than tracking operating hours
for each individual piece of equipment
Response: Todays final rule has been revised to allow the option of performing multiple leak
surveys. Please see the response to comment EPA-HQ-OAR-2009-0923-1014-9. EPA disagrees
with the comment on emission factors. EPAs analysis of the methods used in subpart W is
outlined in the April 2010 proposed rules Technical Support Document (TSD) found in
docket (EPA-HQ-OAR-2009-0923-0027). To provide the data quality necessary to inform
policy, reporters must accurately determine the amount of time of emitting GHGs from each
source.

98.234(a)(1) text clarification: The text in 98.234(a)(1) should be revised to read,
60.18(i)(1) and (2) of the Alternative Work Practice.
Response: EPA agrees and todays final rule has clarified the text.

Organization: ConocoPhillips Company
Commenter: Dan F. Hunter
98.233(u) GHG volumetric emissions
ConocoPhillips Comment:
a.) 98.233(u) requires of onshore petroleum and natural gas production facilities quarterly
sampling of natural gas for GHG content for the following categories: natural gas pneumatic
high bleed device venting, natural gas pneumatic low bleed device venting, natural gas driven
pneumatic pump venting, conventional and unconventional completions and workovers,
reciprocating compressor rod packing venting, well testing venting and flaring, associated gas
venting and flaring, acid gas removal vent, and centrifugal compressor wet seal degassing
1177
venting. 98.233(u)(2)(i) states that the quarterly gas samples must be collected using the
methods set forth in 98.234(b). However, there are no such methods set forth in this paragraph.
We request that EPA allow the use of any consensus based standard for collecting and analyzing
the gas.
Accessibility and Cost of OGI
Some of the comments already submitted on the proposed Subpart W have expressed concern
about the availability and cost of OGI cameras. Accordingly, we thought it appropriate to
provide some cost information so the discussion on this issue is informed with factual data.
A gas detection camera optimized for detecting methane and other VOCs currently costs
approximately $85,000. Another $5,000 should be budgeted to accommodate the training
necessary to make operators competent to find and document leaks in a variety of inspection
scenarios. The costs of this system reflects the unique configuration of sensors that are necessary
to reliably detect gaseous VOCs, including a specially filtered hypersensitive photovoltaic
detector chilled to a temperature of -320F using a compact portable closed cycle cooling system.
It is quite possible, however, that the camera pricing will decline as increased demand creates
production efficiencies and broader adoption draws competitive suppliers into the business.
The Agency accurately points out that many facilities will not need to purchase a camera or train
personnel, and will instead elect to use contractors who have purchased the technology and have
operators with certified training in the use of OGI. Fees charged by contractors for each day of
data collection and reporting generally range from $2,000.00 per day to $5,000.00 per day. The
comments below are typical of the feedback we are receiving from consultants who offer OGI
contract services.
Typical fees are $3,500 per day (12hours of work or $291/hour) which covers all equipment and
crew. Additional charges for reporting would add roughly $500. Travel expenses are additive to
these fees and will vary depending on the ability to access facilities without plane travel or hotel
stays.
The average gas plant takes between 0.5 to 1 day (depending on how many emissions are found).
Note, the EPA threshold of 25,000 tons CO2e/year would be considered an average sized to
small gas plant which would be about 0.5 days or approx $2,250. Large gas plants (approx
230,000 CO2e per year) can be completed in 1.5 days and (approx. 240,000 CO2e/year) in 2
days. Very large gas plants in the range of 350,000 tons CO2e per year with numerous leak
locations can take as long as 3 day.
1178
FLIR Systems believes that sufficient equipment and trained personnel can be made available to
meet the demand from petroleum and natural gas systems once monitoring requirements take
effect in 2011. There are approximately 500 OGI cameras now in service, with approximately
100 additional cameras deployed each year. Contractor services are widely available, with
approximately ten companies in the field now using highly sensitive OGI technology to detect
methane and VOCs. Given that optical imaging surveys need only be taken once a year under the
proposed Subpart W, FLIR Systems believes that there is adequate time over the coming year for
the OGI industry to meet the needs of companies subject to Subpart W.
We have spoken to most of the consultants using FLIR System OGI cameras, and they indicate it
would not be difficult to scale up their operations with additional equipment and trained
operators. Once a service organization has invested in the infrastructure necessary to support the
development of service packages, reporting systems, administration and billing and the
marketing materials necessary to support the service, it is not difficult to add cameras and
operators. Organizations who already offer OGI are typically in the best position to quickly add
competent technicians since they may have labor resources from other diagnostic services that
can be formally trained in OGI. New camera operators will then quickly develop competency
when formal training is augmented by in-field training and insights from trained OGI
technicians.
FLIR Systems manufactures the key camera components, including the detector, read-out
integrated circuits, and cryogenic coolers necessary to build the OGI cameras. FLIR Systems can
scale up production quickly to meet any increase in demand
Other commenters correctly point out that the preamble does refer to customer meters used to
transport natural gas primarily from high pressure transmission pipelines to end users. 180 It is
important to clarify which metered end-customers, if any, will qualify for OGI under the
proposal. We would recommend that all industrial users with metering and regulation and direct
connections to transmission lines be included. Assuming that any customers covered by Subpart
W would be larger consumers, we estimate that there would be between 2 and 25 customers in a
utility like the one surveyed.
Response: EPA has clarified todays final rule and does not include customer meters. Please
180
75 Fed. Reg. at 18,617.

1179

Recommended Phase-In of Certain Underground Pipelines. Although EPA suggests in the
preamble to the proposed Subpart W that underground pipeline fugitive emissions cannot be
detected using OGI, we have some experience with pipeline inspection and the link below
provides an example of how a gas leak can be detected as it leaks through a manhole cover in
downtown Boston.
(http://www.flir.com/thermography/americas/us/OGI/naturalgas/)
This example clearly illustrates that OGI is capable of finding leaks emanating from
belowground pipelines.
Thus, we suggest that EPA consider a phase-in period for use of OGI as a tool for detecting
leaks from underground pipelines. A phase-in would be appropriate for two reasons. First, this
application is the most challenging for OGI. We have found the cameras we deploy for this
application need to engage a special high sensitivity mode to detect gas that may be leaking up
through pavement.
Second, we feel it is premature to require LDCs to use OGI to detect pipeline leaks until more
field work has been completed on OGI applications for pipeline leak detection. We believe that
periodic scanning of manhole covers and other areas may prove effective, but believe that a year
of experience deploying OGI in the LDC targets sited in the table above would be advisable
before specific recommendations on work practices can be formulated and incorporated into
training programs.
Response: EPA disagrees with leak detection of pipelines for todays final rule. EPAs analysis
of pipeline emissions quantification methods is outlined in the Technical Support Document
(TSD) for the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0027), and
in the April 2010 preamble Section II.E.
OGI Sensitivity Thresholds and Performance Standards
The sensitivity thresholds described in 40 CFR part 60 subpart A (i)(1) and (2) require an OGI
camera to detect a methane leak of 60 grams per hour. Contrary to EPAs discussion of the AWP
sensitivity threshold, we have found cameras with this sensitivity level to be suboptimal for OGI.
The cameras utilized today by the natural gas and petroleum sectors, and the cameras produced
1180
for OGI by our company, are able to clearly detect a leak rate of 1 gram per hour of methane or
less. We are not aware of any gas company, petrochemical company or consulting services
company currently using an OGI camera that does not perform to this standard.
Although our company also manufactures lower performance infrared cameras, that are sensitive
in the 60 grams per hour range, we note that cameras operating at this sensitivity level may be
effective in controlled environments but, are highly susceptible to adverse environmental
conditions or operator error. Field dynamics which may include significant wind currents,
varying thermal background conditions and the degree of operator proficiency do not make
these cameras the best option for OGI.
We believe that EPAs reference to the Alternative Work Practice could encourage the use of
infrared cameras that are unable to find leaks in a real world outdoor application. We encourage
the EPA to modify the sensitivity threshold to a level consistent with field validated performance
thresholds to a leak rate of 1 gram per hour of methane or less.
Response: EPA disagrees with this comment. Please see EPAs analysis of the optical gas
imaging instrument in EPAs preamble to the Alternative Work Practice to Method 21, and in
docket EPAHQ-OAR-2003-0199-0005.
Commenter:
General Comments Use of Direct Measurement
On page 18611 of the pre-amble, the EPA states, In this supplemental proposal, EPA is
requiring the use of direct measurement of emissions for only the most significant emissions
sources where other options are not available, and proposing the use of engineering estimates,
emission modeling software, and leak detection and publicly available emission factors for most
other vented and fugitive sources. We disagree that the EPA is requiring use of direct
measurement of emissions "for only the most significant emissions sources where other options
are not available." Direct measurement is required for transmission tanks, wet seal degassing,
reciprocating rod packing venting, flare stacks. Requiring direct measurement of transmission
tanks, wet seal degassing and reciprocating rod packing venting will require field personnel to
visit each site to determine which sources must be reported, and to measure each of those
sources. Yates Petroleum Corporation, and its subsidiaries, is responsible for nearly 4,000 well
sites and hundreds of miles of pipeline. Sending staff to even a portion of these sites will result in
emissions from vehicles, wear and tear on equipment, and valuable time lost by staff in the field
- none of which appears to have been taken into consideration when calculating the cost of
implementing this rule.
Yates requests that EPA reconsider the use of direct measurement for these sources after one
year. If these sources are less than 5% (in alignment with The Climate Registrys deminimus
emissions) of the total GHG emissions after the first reporting year, Yates recommends that EPA
1181
propose emission factors and assumptions as an alternative to direct measurement.

Furthermore, in the Technical Support Document (pp. 28-29), the EPA states, another key issue
is that all measurement technologies discussed require physical access to the emissions source in
order to quantify emissions. Many sources in a producing field (tanks, compressors, etc.) are
located in extremely remote areas and are therefore, by most definitions, inaccessible. Direct
sampling of those field sources will include previously unscheduled trips of environmental staff
which can take staff members off-site for days at a time, traveling in dangerous and remote areas
throughout the course of the year. Direct sampling of these sources is not only burdensome on
our staff, it is dangerous for minimal benefit when emission factors could be developed for such
sources.
Lastly, Yates will rely on 3rd parties to conduct most direct measurement samples. Yates
requests guidance from the EPA on how to reasonably work with 3rd party vendors to ensure
their sampling methodologies meet EPA requirements.
Response: EPA disagrees with this comment. Please see response to comments EPA-HQ
OAR-2009-0923-1060-1 and EPA-HQ-OAR-2009-0923-1015-1. In addition, it is the
responsibility of the reporter to manage any third party vendors they may choose to contract for
work. The designated representative (DR) is the entity that is responsible for submitting the
emissions data pursuant to todays final Rule. Please see the response to EPA-HQ-OAR-2009
0923-1024-16.

Commenter:
75 FR 18620: 1. Direct Measurement
Comment: WBIH recommends limiting the direct measurement of storage tanks at onshore
natural gas transmission compressor station facilities to condensate storage tanks.
EPA stated, "For example, storage tanks in the onshore natural gas transmission segment
typically store the condensate (water, light hydrocarbons, seal oil) from the scrubbing of pipeline
quality gas. The volume and composition of liquid is typically low and variable, respectively, in
comparison of hydrocarbon liquids stored in the upstream segment of the industry. Hence the
emissions from condensate itself in the transmission segment are considered insignificant."
Response: EPA has clarified todays final rule to state that in regard to tank emissions, only
transmission condensate storage tanks require monitoring. Please see the response to comment
EPA-HQ-OAR-2009-0923-1039-20.
1182
Commenter:
75 FR 18612: A. Overview of Proposal, Amendments to the General Provisions
Comment: WBIH recommends the use of direct measurement on manifolded vent lines, existing
and newly constructed, in response to this measurement option, as an optional monitoring
method for emissions quantification for storage tanks at onshore petroleum and natural gas
production facilities.
E&P Tanks is best suited for upstream operations, and for data accuracy, the program requires
site-specific information to determine emission rates. The E&P TANK model allows the user to
input compositional analyses from pressurized oil and gas samples to simulate flash generation
in storage tanks. Specifically, the minimum inputs needed for the model are: separator oil
composition; separator temperature and pressure; sales oil API gravity and Reid Vapor Pressure
(RVP); sales oil production rate; and ambient temperature and pressure. Since separator oil
composition is a key input in the model, E&P TANK includes a detailed sampling-and-analysis
protocol for separator oil. This will require additional costs associated with the required
sampling. Additionally, E&P Tanks is not "available for free" as stated in the Preamble (Section
1I.E.1 (2)). There is a cost of approximately $500 per copy (single-user or single computer
software) for API nonmembers.
There is much on-going discussion in the oil and gas industry about E&P Tanks and the resulting
emissions estimates. When using other emissions estimation models or calculations, such as
Vasquez Beggs Equation, gas-oil ratio (GOR) and throughput of hydrocarbons, and process
simulators (i.e. HYSIM, HYSYS, etc.), there is wide range of emission estimation results, with
the highest estimate being on the order of magnitude three times the lowest estimate.
Response: Please see response to comments EPA-HQ-OAR-2009-0923-1074-34 and EPA-HQ
OAR-2009-0923-1061-10.
Commenter:
EPA seeks comment on whether this alternative approach better estimates annual facility
emissions without resulting in additional reporting burden to the facilities. (page 78)
Detecting, quantifying and repairing small leaks should require minimal burdens, where as
detecting, quantifying and repairing larger leaks should have burdens commensurate with the
leak volume. If larger leaks are present, operators will already have a financial and safety-based
incentive to repair them. The alternative approach is preferred.
Response: EPA will allow multiple leak detection surveys. Please see the response to EPA-HQ
OAR-2009-0923-1014-9.
1183

Commenter:
It is important to allow for flexibility in calculating and reporting of greenhouse gas emissions
for natural gas systems given the nature of such systems. As evidenced by MidAmericans
experience in California, a one-size-fits-all approach to reporting greenhouse gas emissions, even
in a specific industry sector, does not work well. Establishment within the regulation of default
emission factors that overstate emissions without the ability to utilize approved alternatives does
not advance accuracy in measurement and reporting. Having to obtain individual variances or
approvals due to differing circumstances consumes agency and company resources and skews
public perceptions on emissions. EPA should ensure that reporting entities adhere to recognized
protocols that will produce reasonably accurate emissions data rather than prescribing precise
methodologies within the rules themselves.
There are circumstances when differing approaches should be applied to account for greenhouse
gas emissions. For example, for larger leaks in natural gas pipeline operations it is appropriate to
use direct measurement techniques to account for emissions. However, for leaks from
minor/smaller components the use of analytical methods and emission factors are more
appropriate. It is important to give businesses flexibility in selecting the most appropriate
detection and measuring approach to ensure accuracy.
Response: EPA disagrees with the comment that EPA should allow reporters to choose methods
that are not specified in subpart W. Please see The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98) preamble Section L. EPAs analysis of methods is outlined in
the Technical Support Document (TSD) for the April 2010 proposed rule found in docket (EPA
HQ-OAR-2009-0923-0027).
Monitoring methods and emissions quantification for: Fugitive emissions from onshore natural
gas transmission compression
NMGC appreciates that EPA revised Subpart W so that it no longer requires leak detection of all
the 24 originally listed sources and has allowed engineering estimation or component counts to
calculate emissions for several sources. EPA also reduced the number of sources for which a leak
detection survey is required to identify emissions which reduces our burden. However, the new
Subpart W is still labor intensive and in some cases not even possible to calculate using the
proposed methods.
1184
Response: EPA disagrees with the comment on burden. Please see response to comment EPA
HQ-OAR-2009-0923-1099-8 and the preamble Section III.B. The commenter did not provide
sufficient details as to what calculations may be difficult.
Section 98.234(d)(2). The protocol directs the technician to use anti-static wraps or other aids
if the high volume sampler is not able to completely capture all emissions from the source, but
does not specify how such leakage is to be detected or how these aids are to be used.
Response: The commenter has misinterpreted the rule. The rule prescribes that the
manufacturers operating parameters must be followed when operating the leak detection or
measurement equipment.

Paragraph 98.234(1) should be revised to include the phrase of the in the first sentence (in
accordance with 40 CFR part 60, subpart A Section 60.18(i)(1) and (2) of the Alternative work
...) and to delete the word "proposed" in the last sentence ("you must operate the optical gas
imaging instrument to image the source types required by this proposed reporting rule...").
Response: EPA has clarified todays final rule text. Please see the response to comment EPA
HQ-OAR-2009-0923-1039-31. EPA has also replaced proposed reporting rule with the word
subpart in todays final rule.
Fugitive and Vented Emissions
We support EPAs proposed fugitive and vented emission estimating methods that rely on a
combination of leak detection, population count and emission factors. We agree that the methods
proposed are conservative and may tend to overestimate emissions, because the methods assume
that whenever a leak is found it occurred for the entire annual reporting period.
As an alternative, EPA could provide the option for Operators to more accurately estimate
emissions using direct measurement and detailed leak repair reporting methods, where the
1185
Operator can quantify the leak rate, make repairs to stop the leak, and only report emissions for
the period of time that the leak actually occurred, rather than assuming the leak continued for the
entire reporting period.
We do not support EPAs proposal to exclude fugitive emissions from natural gas pipeline
segments between compressor stations, crude oil pipelines, and tank terminals. EPA has
proposed this exemption based on the assumption that fugitive emission leaks are repaired,
whenever found in this industry sector.
This assumption is not always true, and therefore, we request that these fugitive emission sources
be included in the MRR.
Response: EPA disagrees with the comment on direct measurement. Please see the response to
comment EPA-HQ-OAR-2009-0923-1155-12. EPA disagrees with the commenter regarding
onshore natural gas transmission pipelines, crude oil pipelines and tank terminals. EPAs
analysis is outlined in the Technical Support Document (TSD) for todays final rule found in
docket (EPA-HQ-OAR-2009-0923) and the April 2010 preamble Section II.C.
In the 2009 EPA proposed reporting requirements, the EPA limited the reporting threshold to
25,000 metric tons/year per facility based on the cost effectiveness test to capture most of the
GHG emissions while limiting excessive costs. However, according to IPAA analysis, this
proposal will result in 43% of the capital costs to comply will be by the petroleum and natural
gas industry to report an estimated 3% of the nations emissions. While EPA contends that it
underestimated the amount of GHG emissions, IPANM would strenuously object to the use of
any data from the New Mexico Environment Department as a basis for emissions in the State.
The Department has admitted to difficulty in collections and reporting due to staffing and
consulting conflicts which resulted in incomplete figures submitted to meet a deadline for a
Governors taskforce on climate change. While there has been an additional report since the first
submission, the scientific basis for the NMED numbers have not been submitted to industry for
review or discussion.
Response: EPA disagrees that petroleum and natural gas systems account for only 3% of
greenhouse gas emissions. Please see response to comment EPA-HQ-OAR-2009-0923-1005-6.
EPA disagrees with the comment on costs. Please see the Economic Impact Analysis Section 5.
The commenter is not clear as to what data sources they claim EPA used that may be from the
NMED. The sources used for estimating emissions for the threshold analysis are outlined in the
0923)
1186

EPA Should Clarify that 40 CFR 98.233(p)(3) has Two Options vs. One.
Dow comments that it appears that there should be an "or" inserted between the regulatory text in
proposed 98.233(p)(3)(i) and (p)(3)(ii) as the first paragraph calls for the use of an optical gas
imaging instrument and the second paragraph calls for direct measurement of the emissions
using a high flow sampler or a calibrated bag. It appears that these two paragraphs are
independent options, thus an "or" should be inserted between them.
Response: The commenter has misinterpreted the rule. EPA disagrees that an or should be
inserted between the leak detection and leak measurement paragraphs in the monitoring
description for reciprocating compressors. The leak detection in the first paragraph is to identify
equipment leaks; the second paragraph is to measure those identified leaking emissions. Both
are required for compliance with subpart W.
Second, the proposal would impose detailed emissions estimating and measurement
requirements that will be far more costly than EPA predicts and would produce information that
is no more useful to the Agency than information that could be gathered in significantly less
prescriptive and costly ways. For example, emissions from all tanks must be calculated, even
tanks that have small throughput and, therefore, nominal emissions. Similarly, although direct
measurement is required for only a few source types, the testing would be prohibitively
expensive and would produce information that is of questionable validity given that it would Air
and Radiation Docket and Information Center represent only a point in time, which does not
necessarily shed light on performance over an entire year.
Response: Todays final rule requires direct measurement for sources that are determined to
have large uncertainty, and for which all other methods significantly underestimate emissions.
EPA requires such direct measurement to produce the necessary data quality to inform future
policy. Please see the Technical Support Document (TSD) for todays final rule found in docket
(EPA-HQ-OAR-2009-0923). EPA has revised several of the monitoring methods in todays
final rule to reduce burden. Please see the preamble Section II.F. EPA disagrees with the
comment on costs. Please see the Economic Impact Analysis Section 5.

1187

Offshore LNG facilities consist of the LNG carrier with its integral LNG tanks, pumps, and
vaporizers, and the mooring and sendout buoy and underwater pipeline. As is true for onshore
facilities, specialized installed equipment is designed, operated, and maintained to contain the
fluid; installed detection monitors each area and alarms if an emission is detected; personnel
inspect and control the processes and will immediately stop an operation in the event of a leak, to
protect the carriers structures from the cryogenic fluid; and emergency shutdown valves provide
ultimate control to stop the process. Each ship moors for only long enough to vaporize and
offload its cargo. The facility operates only when an LNG carrier is at the deepwater port.
Accordingly, the requirement for annual leak detection would be inapplicable. Further, the U.S.
Coast Guard thoroughly inspects the safety systems of each vessel upon each arrival, and would
prevent mooring, or stop an offloading, in the event of a deficiency or leak.
Response: EPA disagrees with the comment, and is including the LNG facilities in todays final
rule. Please see the response to comment EPA-HQ-OAR-2009-0923-1025-1 and EPA-HQ-OAR
2009-0923-1151-130.
Leak Detection and Reporting
The proposal for leak detection and reporting would impose significant additional costs and
would result in an overstatement of the actual level of facility emissions. The proposed rule
would require utilities to conduct annual leak detection at above grade M&R stations and to
NGA Comments on EPA Subpart apply emission factors to the leaking sources. While current
practices require regular leak detection, and effective methods such as bubble testing are used,
the proposal specifies optical imaging, a method that appears to be less accurate. Optical imaging
requires expensive monitoring equipment, significant time for conducting measurements in the
field, and extensive recordkeeping. (On the other hand, current methods using soap solutions,
although not quantified, are accurate in pinpointing leaks. The simplicity of this method allows
all field personnel to find and fix leaks immediately.) A simple and effective method should be
the preferred choice. We believe that the value of optical imaging needs to be demonstrated and
defined before any mandate is imposed.
HQ-OAR-2009-0923-1299-18 and the preamble Section III.B. EPA has revised todays final
rule and will allow specific additional leak detection methods. Please see the preamble Section
II.F.
1188

We also question whether there would be enough equipment or trained operators to conduct such
annual surveys at all the M&R stations that would be affected. Also, when utilities conduct leak
detection, they promptly repair leaks that can be readily repaired. Reporting the repaired
components as if they were leaking for the entire year will overstate actual emissions. We would
respectfully ask that EPA first investigate the accuracy of such programs in terms of the
methodology for emissions quantification before imposing and implementing this procedure, and
also undertake a comparative assessment of other existing programs and methods to determine if
there is a more accurate and cost-effective alternative.
Response: Concerning the availability of equipment or trained operators, for certain sources
preamble Section II.F. EPA has revised todays final rule to include multiple leak detection
surveys to account for repairs. Please see response to comment EPA-HQ-OAR-2009-0923
1014-9. Regarding the comment on assessment of alternative programs, EPA detailed its
research into the prescribed methodologies and alternatives for all sources. Please see the
0923).

As noted by the EPA, the natural gas industry is taking a proactive role in identifying natural gas
leaks and then repairing the identified leaks. As a result of the extra leak detection imposed by
this rule, further leak repairs will be performed. DTE Energy supports calculating emissions in
which the leak time is measured from the time the leak was identified to when the leak was
repaired, or to the end of the year. This will give a more accurate quantification of fugitive
emissions that factor in the reductions resulting from leak repairs.
Response: EPA has revised todays final rule to include multiple leak detection surveys to
account for repairs. Please see response to comment EPA-HQ-OAR-2009-0923-1014-9.

Measurement Options: Three emissions quantification options were evaluated. These options are
shown below.
1189
1190
1191
Discussion of measurement options:

3) Emission Factors.
In many cases manufacturers data for equipment bleed rates is available. There are several
issues which suggest that significant errors in estimated emissions may result when one uses
OEM data. First, there is no industry standard concerning the reporting of instrument bleed rates
and thus manufacturers report information in a wide range of units and under varying operating
conditions. In addition the data reported by manufacturers has not been independently verified.
USEPA has found large discrepancies between OEM bleed rate data and actual field data.
Factors not reflected in available OEM data, such as gas pressure and maintenance history,
significantly influence emissions rates. While the American Petroleum Institute API (2009)
states that the use of manufactures data is the most rigorous approach the Compendium also
acknowledges that manufactures emission rates tend to be lower than emissions observed.
Emission factors (EFs) for pneumatic devices have been developed. Many of these EFs were
published in the 1997 GRI/USEPA Report, Methane Emissions from the Natural Gas Industry.
The API Compendium compiles these and other EFs (see Table 5-15, pages 5-68 and 5-69) and
estimates uncertainties in some cases. Where specified, reported uncertainties range from +/-33%
to +/-407%. This indicates that use of EFs would result in very unreliable estimates of vented
methane emissions.
4) Direct measurement:
Actual site specific measurement of vented emissions from low and high bleed pneumatic
control devices is accepted to be an accurate method to quantify methane (and CO2 if present in
the gas) emissions. One may characterize emissions from each pneumatic device at a facility
using a bagging technique (or other method) where emissions from the device are captured and
the volume of released gas is measured. Gas analysis then allows one to calculate actual CH4 and
CO2 emissions. This technique is time consuming, labor intensive and expensive. In addition
emissions may subsequently change as the result of factors such as maintenance activities and
gas pressure changes, necessitating additional measurement.
5) Meter Instrument Gas Consumption:
Actual metering of instrument gas consumption and periodic measurement of gas composition
will also provide an accurate determination of GHG emissions. Modifications to instrument gas
plumbing and installation of one or more gas meters will be required initially and this will result
1192
in upfront material and labor costs. However, reporting in subsequent years will be very simple
and easy, especially given the fact that periodic gas analysis will be required for other GHG
emission calculations (e.g. stationary combustion emissions). In addition, changes in system
operating conditions (e.g. line pressure, maintenance activities, instrument modifications)
designed to reduce emissions will immediately be reflected in the volume of instrument gas
consumed. Facility operators receive immediate feedback on their efforts to reduce emissions
and can monitor instrument gas consumption in real time.
Response: EPA disagrees with the comments regarding direct measurement of pneumatic
device venting or metering pneumatic device gas consumption. EPA reduced burden while
maintaining the necessary data quality to inform future policy. EPA has revised the final rule to
allow the use of emission factors for pneumatic devices and to allow reporters to complete a
count of total pneumatic devices in three years, with best available data being acceptable for
years one and two if the count is incomplete to reduce burden while maintaining the necessary
data quality to inform future policy. Furthermore, EPA allows reporters to update the total count
of pneumatic devices based on changes in the system beyond year three of reporting. Please see
the preamble Section II.F and response to comment EPA-HQ-OAR-2009-0923-0055-11. EPA
disagrees with frequent sampling of gas composition as this would create additional burden,
please see the preamble Section II.E.

Proposed Rule Section 98.233(q)(3) Leak detection and leaker emission factors for Onshore
natural gas transmission compression facilities.El Paso requests confirmation from the EPA
that there are no monitoring requirements for centrifugal compressor dry seals from units
operated at natural gas transmission facilities.
Response: Reporters are to monitor emissions through the centrifugal compressor dry seal vent
line under the source centrifugal compressor venting. Please see response to comment EPA
HQ-OAR-2009-0923-1011-46.

Organization: Red Cedar Gathering
Commenter: Ethan W. Hinkley
Direct Measurement of Compressor Rod Packing
Under the proposed rule EPA is requiring direct measurement of reciprocating compressor rod
packing and blowdown valves for three operating modes. One of these is for a compressor that is
not-operating and depressurized. Requiring measurements for this mode will require unnecessary
1193
down-time, loss of revenue, and unnecessary emissions being vented to atmosphere for the
purpose of quantifying an insignificant source of emissions. When a compressor is not operating
the unit is isolated from the gas flow, therefore no gas will flow to the rod packing and
blowdown valves. Other emission regulations under the Clean Air Act specifically state that it is
not required to start an engine solely for the purpose of conducting an emission test, nor should it
be required to shutdown, and blowdown, and engine solely for the purpose of conducting an
emission test. Red Cedar strongly recommends that EPA remove the requirement for leak
detection and measurement for compressors in a non-operating, depressurized mode.
Response: EPA does not require the shutdown of compressors to collect data required under
subpart W. Please see Section II.F of the preamble to todays final rule. EPA has revised
todays final rule to allow compressor venting measurement in the as found mode and that
shutdown depressurized mode must be measured at least once every three years for each
compressor. Please see response to comment EPA-HQ-OAR-2009-0923-0055-16.
Aka recommends that the reporting requirement, (98.236(c)(18)(i)) to measure throughput for
each compressor be deleted. Alternatively, if EPA determines that compressor throughput is
needed, that information should be provided on a per facility basis (i.e., as an aggregate for all
compressor at a given facility rather than for each compressor individually) and throughput
should be based on engineering estimates rather than direct measurement.
Response: EPA disagrees that average annual compressor throughput should be eliminated.
Please see the response to comment EPA-HQ-OAR-2009-0923-1039-22. EPA has clarified that
compressor throughput meters are not required, only annual throughput using engineering
estimation based on best available data. Please see the preamble Section II.F.
There is an important inconsistency between the preamble and the proposed rule text. The
preamble addresses monitoring methods and emissions quantification for large compressors (see
Table W-4). In contrast, the proposed rule text would encompass all reciprocating compressors,
regardless of size. IPAMS requests that EPA modify the rule to be consistent with the stated
intent in the preamble by stating that emissions should be reported only for large compressors,
which should be defined as compressors of 2,000 hp or more.
Response: EPA considered these comments, and has revised the rule to allow in onshore
production for both reciprocating and centrifugal compressors, that population emission factors
can be used. Please see response to comment EPA-HQ-OAR-2009-0923-1063-1.
1194

Best Available Monitoring Methods
We agree with EPAs proposed decision not to allow the use of best available monitoring
methods for part or all of the first year of data collection. EPA has already failed to issue timely
final greenhouse gas reporting requirements for petroleum and natural gas systems as prescribed
by Congress, and these facilities should be required to promptly and fully comply with the
monitoring methods established under the proposed rule.
Response: EPA disagrees with the comment regarding not allowing best available monitoring
methods. Please see the preamble Section II. F.
Implement a phased approach of the use of best available monitoring methods for source types
requiring metered flow rates or monitored parameters.
Response: EPA disagrees with the comment regarding a phased approach for best available

To address the amount of work and resources necessary to implement the rule as proposed, EPA
should:
* Enable the use of BAMM for source types requiring metered flow rates, extensive sampling, or
monitored parameters for the initial year of 2011 along with provisions to request extensions of
BAMM beyond the end of 2011. This is particularly important for a sector like oil and gas where
operations are spread out over a basin and are not staffed at all times. Often there is little existing
infrastructure at these disparate locations and systems for monitoring and gathering information
will likely need to be implemented.
- In addition, given that many of the sources are relatively minor individually and can be grouped
1195
together for estimation purposes, several simplifications that do not overly compromise data
quality are detailed later in these comments.
Response: EPA will allow the use of best available monitoring methods for some sources.
Please see the preamble Section II.F.

Use of Best Available Monitoring Methods
EPA requested comments on the proposal on the use of Best Available Monitoring Methods
(BAMM) for part or all of the first year of data collection. DTE Energy agrees that the BAMM
methodology of Section 98.3(d) is not well suited for fugitive leak detection and measurement
when the reporting company is only required to perform a single annual test. On the other hand,
BAMM for fugitive leak detection and measurement may need to be available for the entire year
and not just for a 3 month time period in order to meet monitoring and reporting requirements
during the first year of implementation under a final rule.
DTE Energy is concerned that if optical imaging equipment remains the only allowed method for
fugitive leak detection, there will not be enough leak detection equipment available in the first
year. If a best faith effort was made to obtain the equipment and train staff to use the equipment,
DTE Energy believes that flexibility should be allowed in completing first year leak detection
and measurement requirements. Unlike requirements to install fuel flow meters (i.e., to comply
with the General Stationary Fuel Combustion Sources requirements under Subpart C), companies
may not know whether they will need extra time to complete the leak detection and monitoring
requirements under Subpart W until a substantial part of the first year has passed. DTE Energy
supports the utilization of BAMM for the entire first year for instances when companies can
objectively show they made best faith efforts to complete the leak detection and measurement
required under Subpart W. DTE Energy also believes that a BAMM time frame similar to that
allowed under 98.3(d) should be allowed for the first three months of 2010 if any continuous
measuring devices are required to be installed under Subpart W.
12.2
FREQUENCY OF MONITORING
Commenter:
1196

Annual Reporting
What is the rationale for annual reporting when experience with the U.S. Inventory of GHG
emissions show very gradually changes in the total inventory on a year to year basis? For
instance, even with the recent recession total inventory was only down about 2 to 3%. Has EPA
given any thought to whether the large financial burden on industry should be required each and
every year when a biannual report could suffice?
Response: EPA has not changed the frequency of reporting. Please see The Final Mandatory
GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.H.
Annual reports from small oil and gas businesses operating marginal wells are excessive. As
previously stated, marginal wells are mature crude oil and natural gas producing properties that
have lost their initial, high production rates and instead operate on the much lower, flat end of
the natural production decline curve. Emissions from these types of wells will not change
significantly from year to year. At a minimum, EPA should reconsider this requirement. We
recommend reporting from small oil and gas operators operating marginal wells be no sooner
than every 5 years.
Response: EPA has revised todays final rule monitoring requirements, decreasing the burden
on small business. Please see the preamble Section III.D. EPA has not changed the frequency of
reporting. Please see The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part
98) preamble Section II.H.

General Comments Natural Gas Composition and Sampling Frequency
In the Technical Support Document (Page 44), the EPA states that, these gas composition
estimates are assumed to be available with facilities. But this may or may not be a practical
assumption. In the absence of gas composition, periodic measurement of the required gas
composition for a speciation of natural gas mass emissions into CH4 and CO2 could be a
potential option. For facilities required to monitor fuel, PAW has found that the content of
natural gas remains relatively constant by producing field. It is unnecessary to sample gas semi
annually or for every site, as the content changes over years, not months and largely not by
1197
basin. Monthly gas samples are taken in fields that are still under development until the field
stabilizes. Once the field stabilizes, those samples are generally taken no more frequently than
semi-annually. Once a wellhead stabilizes and is tied into a gathering system, it is not industry
practice to sample the fuel more frequently than semi-annually.
PAW recommends that the EPA allow companies to collect representative gas samples by basin
rather than site-by-site for sites required to do quarterly or semi-annual gas sampling as required
by 98.233(d)(1) and 98.233(u)(2)(i).
Response: EPA agrees, and has revised todays final rule regarding onshore production gas
composition data collection. Please see the preamble Section II.E and response to comment
EPA-HQ-OAR-2009-0923-1060-11.

In 98.233(p)(4) CAPP would like clarification on how often these measurements will be
required. The emission flow rate will change as the packing wears. Implementing a
representative testing program would produce a large data gathering and retention burden.
Additionally CAPP requests that the EPA provide guidance on how to handle situations where
the unit may only be down due to upset and there is insufficient time to organize and conduct the
required measurements.
Response: EPA has revised todays final rule regarding compressor monitoring. Please see the
preamble Section II.F and response to comment EPA-HQ-OAR-2009-0923-1011-44.

For venting not associated with well work (i.e., compressor rod packing, compressor wet seals,
pneumatic devices and pumps, amine unit stacks), quarterly sampling will produce little to no
variability in the results when the source of gas is the same reservoir from quarter to quarter. We
thus request the option to perform the sampling every two years consistent with other sections of
the rule (e.g., 98.233(f)(1)(i)(C)).
Response: EPA has revised todays final rule to allow the gas analysis for produced gas in
98.233(u) to use the reporters most recent gas composition based on available sample analysis
of the field. Please see the preamble Section II.E.
1198

Commenter: Tisha Conoly Schuller
A second concern is the frequency of the sampling of emissions which we believe is redundant,
costly, and adds little new data in the calculation of the emissions.
Response: EPA has revised todays final rule regarding onshore production gas composition
data collection. Please see the preamble Section II.E.
Commenter:
WBIH recommends language which provides the ability to cease leak surveys or reduce the
frequency of leak surveys.
The proposed rule does not have any language to reduce or cease leak surveys.
Annual leak surveys should be phased-out as improved equipment- or component based emission
factors are developed and data accuracy objectives are met.
Response: EPA disagrees with less than annual reporting. Please see The Final Mandatory
GHG Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.H. EPA has
provided provisions for facilities to cease reporting. Please see The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98) preamble Section II.H.

Commenter:
EPA seeks comments on whether, if implemented, multiple surveys should be optional or
required for owners or operators. (page 78)
Companies should be required at a maximum to conduct one annual survey. Alternative less
frequent surveys should be considered based on the greenhouse gas leakage frequency and rate
as documented by the company for each facility.
Response: EPA disagrees and requires at least an annual survey, but will allow multiple leak
detection surveys. Please see response to comment EPA-HQ-OAR-2009-0923-1014-9. EPA
disagrees with less than annual reporting. Please see The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98) preamble Section II.H.
1199

Frequency of the leak survey (page 18623)
NMGC feels the leak survey, if required, should be conducted annually as proposed in the rule.
However, the rule should allow a company to voluntarily conduct surveys on a more frequent
basis in order to be able to report emissions from a shorter time period than a year.
Response: EPA will allow multiple leak detection surveys. Please see response to comment
EPA-HQ-OAR-2009-0923-1014-9.

Section 98.233(k) Transmission storage tanks. Section 98.233(k)(1) and (2) requires the flow
rate of vented emissions from transmission storage tanks in onshore natural gas transmission
compression facilities where the vapors are not routed to a vapor recovery or a thermal control
device to be determined using either optical imaging or, if the vapors are continuous, a flow
meter. Section 98.233(k)(3) specifies how to calculate emissions where the vapor is sent to a
flare. Section 98.233(k)(1) refers to Section 98.234(a)(1) for how the optical gas imagining is to
be conducted. Section 98.234(a) says You must use the method described as follow to conduct
annual leak detection of fugitive emissions from all source types listed in Section 98.233(p)(3)
(i) and (q) Neither Section 98.233(k)(1) or Section 98.234(a) indicates the frequency for
performing the optical gas imaging for transmission storage tanks. API assumes the optical gas
imaging is to be performed annually and not every time there is a release from the scrubber
dump valve. EPA should clarify the frequency in the rule.
Response: EPA has clarified todays final rule that transmission storage tanks are to be
monitored only once annually. Please see response to comment EPA-HQ-OAR-0923-2009
1024-30.

W14. (Preamble pp. 77-78) EPA considered, but did not propose, requiring a facility to conduct
multiple surveys and to report emissions using the appropriate leaker factors. EPA seeks
comment on whether this alternative approach better estimates annual facility emissions without
resulting in additional reporting burden to the facilities. Further, we seek comment on whether, if
implemented, multiple surveys should be optional or required for owners or operators.
1200
API is not in favor of mandating multiple surveys, as this will result in an excessive burden.
However, a reporter should have the option of utilizing data from multiple surveys, if data are
available.
EPA-HQ-OAR-2009-0923-1014-9.
Flare Emissions
We recommend that all gas that is routed to flare devices be metered. We agree with WCIs
recommendation to increase the frequency of gas composition sampling, creating more accurate
emission estimates.
Response: EPA disagrees that all gas that is routed to flare devices must be metered, as this data
quality is not necessary to inform future policy, and would create additional burden. EPA
disagrees with the comment on gas composition sampling. Please see response to comment
EPA-HQ-OAR-2009-0923-1042-23.

Reduced frequency of measurements or process sampling, especially for invariant parameters.
Quarterly sampling is required for numerous parameters such as field gas, AGR process gases,
and hydrocarbon and water storage tank liquids. EPA has not provided or supported the technical
basis for requiring this frequency. In addition, the costs are excessive as demonstrated above and
should be reduced to an annual basis or less.
Data gathered as part of this process should inform the correct sample frequency over time and
the sampling frequency should be relaxed when it is demonstrated that the process parameters
are not significantly varying over time. For example, if two consecutive gas samples show that
the composition is varying by less than a threshold (e.g. change in methane content and carbon
content is less than 10%), then sampling should be relaxed to every other period and then to
every fourth period and so forth unless consecutive samples exceed the threshold. For sampling
and measurement activities required annually (e.g. separator oil and water samples for storage
tank emission estimates), similar criteria to relax sampling to bi- or tri-annual should be
instituted.
1201
Response: EPA agrees and has revised todays final rule regarding onshore production gas
composition data collection. Please see the preamble Section II.E. EPA has not changed the
frequency of reporting. Please see The Final Mandatory GHG Reporting Rule (Final MRR),
(40 CFR part 98) preamble Section II.H.
Reduced frequency of measurements or process sampling, especially for invariant parameters.
Response: EPA has revised todays final rule regarding onshore production gas composition data
collection. Please see the preamble Section II.E.
The section on GHG volumetric emissions generally requires the use of continuous gas analyzer
or quarterly gas analysis for facilities operated by GPA member companies. There is no
justification for quarterly gas analysis. An annual analysis would be sufficient.
Response: EPA agrees and has revised todays final rule regarding onshore production gas
composition data collection. Please see the preamble Section II.E.

Also, when utilities conduct leak detection, they promptly repair leaks that can be readily
repaired. So reporting the repaired components as if they were leaking for the entire year would
end up overstating actual emissions. Were concerned about that as well.
EPA-HQ-OAR-2009-0923-1014-9.
12.3
USE OF METHOD 21 DETECTION METHODS

1202

The monitoring methods specified in proposed 98.234, "Monitoring and QA/QC
Requirements," should be broadened. Section 98.234 would require the use of an
Optical Gas Imaging Instrument for fugitive emissions detection on an annual basis. Essentially,
this would involve screening equipment with an infrared camera to identify and count leaking
components. Although useful for visualizing fugitive emissions, the cost of an infrared camera
set up to visualize equipment leaks is high and training and expertise are required to optimize its
use. Images may be affected by background, ambient temperature, and other factors, rendering
the camera ineffective. Other methods are available for leak screening (e.g., acoustical
instruments, soap bubbles, olfactory, audio, and visual means). A common method for
monitoring fugitive organic emissions from leaking equipment is to employ a portable monitor
using EPA's Method 21 for determination of volatile organic compound leaks. Generally, if a
leak is detected, even with the optical monitor, a Method 21 monitor may be used and is often
required for leak checking equipment after it is repaired under a Leak Detection and Repair
Program. The use of an optical imaging instrument is offered as an Alternative Work Practice for
Monitoring Equipment for leaks in 40 CFR Part 60, Subpart A, 60.18(i)(I) and (2).
Because other alternatives are available, the monitoring methods required under 98.234 should
be revised to include the use of Method 21 fugitive monitoring, especially if such a program is
already in place and readily available.
Response: EPA agrees that specific alternatives to optical gas imaging cameras are acceptable
for the purposes of leak detection for subpart W; please see Section II.F of the preamble.
12.4
USE OF BEST AVAILABLE INFORMATION

The proposed effective date is impractical. It appears EPA is targeting publication of the
finalized Subpart W in late 3rd quarter or early 4th quarter 2010, with an effective date of
January 1, 2011. If the rule is finalized as proposed, there will be certain requirements (e.g., the
acquisition and installation of certain monitors or information systems) that will not be practical
to implement prior to January 1, 2011, and adequate time will not be allowed for the regulated
community to do basic planning for implementation of the new rule. To alleviate such problems,
Anadarko recommends EPA allow use of best available monitoring methods (BAMM) for
emissions from calendar year 2011. Alternatively, EPA should develop a phased implementation
schedule for Subpart W such that the effective date of Subpart W varies by industry segment,
with segments with less burdensome requirements reporting sooner and those with greater
burdens reporting later. We recommend that a three year phase-in period would be appropriate
and that the upstream sector should be included in the last phase.
1203
Response: EPA will allow the use of best available monitoring methods for some sources and is
allowing some data to be developed over a three year period, but is not agreeing to an across the
board phase-in period. Please see the preamble Section II.F.
(Preamble p. 23) EPA seeks comment on the proposal not to allow use of best available
monitoring methods for part or all of the first year of data collection. Further, if commenters
recommend that EPA allow the use of best available monitoring methods for a designated time
period (e.g., three months), EPA seeks comments on whether requests for use of best available
monitoring methods should only be approved for parameters subject to direct measurement, or
also in cases where engineering calculations and/or emission factors are used.
A phased implementation of the rule over a period of several years will be necessary to fully
understand the requirement of the rule, inventory all sources subject to the rule, and develop data
collection systems, etc. to comply with the rule. Best available monitoring methods (BAMM)
should be allowed where flow meters or pressure monitoring complying with Section 98.234(b)
is required. The proposed compliance dates are unrealistic for the vast number of sources subject
to Subpart W that have not been previously subject to regulatory reporting requirements. It is
unreasonable to expect the regulated community to initiate procurement and installation of
expensive monitoring equipment based on the proposed rulemaking. Regulated entities should be
provided sufficient time between todays final rule (expected October 2010) and the
implementation date for the required monitoring systems (January 1, 2011) to allow for detailed
designing and engineering, equipment procurement, installation and calibration, personnel
training, and systems integration. API recommends a phased in approach for monitoring systems
where direct measurement is required similar to the provisions in Section 98.2(d).
Given the geographic dispersion of oil and gas facilities, installation of any additional equipment
or monitors can be very challenging. Thus, API does not agree with EPAs comment in the
preamble that the proposed monitors are widely available and are not time consuming to
install. The inherent nature of the many unstaffed facilities would make installation and
continued maintenance of any additional equipment arguably equally or more challenging for
these facilities relative to a refinery or a chemical plant. API believes that a multi year BAMM
approach is essential to a smooth implementation of this rule.
In addition, API is interpreting the rule to mean that any measurement that is to be collected at a
frequency of once every two years must be collected by the end of the second year of the rule
(December 31, 2012) to allow for the preparation of the 2012 report. If this interpretation is
consistent with EPAs intent, API is fine with that the intent. If there is a requirement to have the
monitoring completed sooner (i.e., December 31, 2011), then API would like to add that to the
list of monitoring activities eligible for BAMM extensions.
1204
Response: In developing the petroleum and natural gas sector initial implementation strategy,
EPA considered using a phased-in approach. EPA determined that a blanket phased-in approach
is not necessary because not every source will be too difficult to monitor in the first year of
implementation. However, EPA considered that it may not be feasible for all reporting entities
to fully comply with 2011 reporting requirements and therefore will allow the use of best
available monitoring methods for certain sources. Please see the preamble Section II.F With
regard to collecting data at a frequency of once every two years, it was not EPAs intent in the
proposed rule to allow reporters to begin this cycle at the end of the second year. EPA requires
that reporters collect this data such that it can be submitted in their 2011 report; todays final rule
stipulates that data which is not collected annually must be collected in the first year. Please see
todays final rule Section 98.233(f) and 98.233(g).
Commenter: Kim Dang
To account for the short lead time and the limitations on availability of compliance resources,
EPA should allow the use of best available monitoring methods for at least the first year of
compliance with Subpart W.
Response: EPA will allow the use of best available monitoring methods in certain cases. Please

Commenter: Kim Dang
Additionally, consistent with our comments on the safety implications of Subpart W in Section
II.A, Kinder Morgan recommends that EPA allow the use of an emission factor, emission rate, or
other best available monitoring method for unsafe-to-monitor or difficult-to-monitor vents at
reciprocating compressor rod packing venting. Facilities can use leak detection methods outlined
in SECTION 98.234(a) to determine if the reciprocating compressor rod packing is venting. For
those rod packings determined to be venting, emissions can be estimated utilizing an emission
rate drawn from Natural Gas STAR databases or other appropriate sources. 181
181
A recommended emissions rate for reciprocating compressor rod packing venting is provided at EPAs Natural
Gas STAR website at http://www.epa.gov/gasstar/tools/recommended.html.
1205

The proposed effective date is impractical.
It appears EPA is targeting publication of the finalized Subpart W in late 3rd quarter or early 4th
quarter 2010, with an effective date of January 1, 2011. If the rule is finalized as proposed, there
will be certain requirements (e.g., the acquisition and installation of certain monitors or
information systems) that will not be practical to implement prior to January 1, 2011, and
adequate time will not be allowed for the regulated community to do basic planning for
implementation of the new rule.
To alleviate such problems, AXPC recommends EPA allow use of best available monitoring
methods ("BAMM") for emissions from calendar year 2011. Alternatively, EPA should develop
a phased implementation schedule for Subpart W such that the effective date of Subpart W varies
by industry segment, with segments with less burdensome requirements reporting sooner and
those with greater burdens reporting later. We recommend that a three year phase-in period
would be appropriate and that the upstream sector should be included in the last phase.
Response: EPA disagrees with the comment regarding a phased-in approach. However, EPA
has determined that for specified emissions sources for certain industry segments, some reporters
may need more time to comply with the monitoring and QA/QC requirements of Subpart W by
January 1, 2011. EPA will allow the use of best available monitoring methods for some sources.
To avoid the issue of an unreasonable schedule set in the 2009 final rule, at a minimum, EPA
should allow the use of Best Available Monitoring Methods (BAMM) for 2011. In addition,
BAMM should include cases where engineering calculations and/or emission factors are used to
quantify emissions.
(Preamble p. 23) EPA seeks comment on the proposal not to allow use of best available
monitoring methods for part or all of the first year of data collection. Further, if commenters
recommend that EPA allow the use of best available monitoring methods for a designated time
period (e.g., three months), EPA seeks comments on whether requests for use of best available
monitoring methods should only be approved for parameters subject to direct measurement, or
1206
also in cases where engineering calculations and/or emission factors are used.
ConocoPhillips comment:
To address the unprecedented amount of work and resources necessary to implement the rule as
proposed, ConocoPhillips requests the use of BAMM for at least the first year for several source
types requiring metered flow rates or monitored parameters and for cases where engineering
calculations and/or emission factors are used to quantify emissions. BAMM is particularly
important for a sector like oil and gas. Unlike a chemical plant or a refinery, oil and gas
operations are spread out over a basin and are often not staffed at all times. Thus, often there is
little existing infrastructure at these disparate locations as opposed to a refinery or a plant.
Therefore, the use of appropriate meters and the counting of pieces of equipment will be harder
for these spread out oil and gas sources. Another very important point is that many companies
will rely on contractors to carry out the work like optical imaging and quantifying emissions
from equipment like compressor wet seals. There may not be a sufficient number of contractors
to serve all the companies who will call on them in 2011 resulting in some companies facing
potential compliance issues. For this reason, BAMM is very important in that it will allow
industry to develop this resource for use in 2012.
Use of Best Available Monitoring Methods
The Clean Energy Group has concerns with the cost of implementing the proposed monitoring
methods and the need to develop procedures and train staff, especially in the timeframe in which
they are proposed to be implemented. Similar to the final mandatory reporting rule promulgated
in October 2009, EPA could permit the use of best available monitoring methods for three
months for parameters subject to direct measurement, and for fugitive leak detection.
Alternatively, as noted above EPA could require simpler reporting methodologies for several
years as it works with the industry to improve methods, and emission factors.
Allow the use of Best Available Monitoring Methods for the first quarter of 2011.
1207
Response: EPA will allow the use of best available monitoring methods for some sources;
Best available methods currently employed by industry. EPA should provide for allowances to
apply current industry standards or best practices. These include, but are not limited to: emission
factors and estimation methods from the API Compendium or other GHG reporting reference
documents, HYSIS and other process simulators, EPA and/or industry initiatives to collect data
and develop better emission factors, and other industry standard approaches. Flexibility in
selecting and applying emission estimation methods is discussed in greater detail in Comment
IV. Industry has developed and used a multitude of emission estimation tools and methods to
develop GHG emission inventories. Many of these are equivalent to or more rigorous than
methods prescribed in the proposed rule. These methods should be allowed provided they are
documented in the monitoring plan.
Response: EPA disagrees with the comment to allow reporters to choose methods not
prescribed in the rule. Please see response to comment EPA-HQ-OAR-2009-0923-1151-9.
To alleviate some of this burden, Noble recommends that emission estimation flexibility - that is,
alternative best available emission estimation methods - be allowed. Although these methods
may differ from those prescribed in Subpart W and Subpart C, their basis would be standard
industry GHG protocols, such as the API Compendium, and company-specific data collection
and engineering calculations developed for on-going annual inventory programs. To ensure data
quality, the alternative methods would be documented in the Monitoring Plan.
Response: EPA disagrees with the comment to allow reporters to choose methods not
prescribed in the rule. Please see response to comment EPA-HQ-OAR-2009-0923-1151-9.
EPA seeks comment on the proposal not to allow use of best available monitoring methods for
part or all of the first year of data collection. Further, if commenters recommend that EPA allow
1208
the use of best available monitoring methods for a designated time period (e.g., three months),
EPA seeks comments on whether requests for use of best available monitoring methods should
only be approved for parameters subject to direct measurement, or also in cases where
engineering calculations and/or emission factors are used.
Best available monitoring methods (BAMM) should be allowed where flow meters or pressure
monitoring complying with 98.234(b) is required. The proposed compliance dates are unrealistic
for the vast number of sources subject to Subpart W that have not been previously subject to
regulatory reporting requirements. It is unreasonable to expect the regulated community to
initiate procurement and installation of expensive monitoring equipment based on the proposed
rulemaking. Regulated entities should be provided sufficient time between the final rule
(expected October 2010) and the implementation date for the required monitoring systems
(January 1, 2011) to allow for detailed designing and engineering, equipment procurement,
installation and calibration, personnel training, and systems integration. BP recommends a
phased in approach for monitoring systems where direct measurement is required similar to the
provisions in 98.2(d). BP recommends for Subpart W the BAMM special provisions in
98.2(d)(1) apply through all of 2011 with the ability to apply for an extension to use BAMM for
a longer period according to 98.2(d)(2).
In addition, where the installation of required metering and/or monitoring equipment would
require a shut down of a facility or unit BAMM should be allowed until the next scheduled
shutdown where the equipment could be installed.
Given the spread out nature of oil and gas facilities, installation of any additional equipment or
monitors can be very challenging. Thus, BP does not agree with EPAs comment in the preamble
that the proposed monitors are widely available and are not time consuming to install. The
inherent nature of the many unstaffed facilities would make installation and continued
maintenance of any additional equipment arguably equally or more challenging for these
facilities relative to a refinery or a chemical plant. BP believes that a first year BAMM approach
is essential to a smooth implementation of this rule.
In addition, BP interprets the rule to mean that any measurement that is to be collected at a
frequency of once every two years must be collected by the end of the first year of the rule
(December 31, 2011) to allow for the preparation of the 2011 report. If this interpretation is
consistent with EPAs intent, BP is fine with that the intent. If there is a requirement to have the
monitoring completed sooner, then BP would like to add that to the list of BAMM extension
request-eligible monitoring activities.
Response: EPA disagrees with the comment that the monitoring methods in todays final rule
are too challenging, as todays final rule does not require the installation of any permanent
monitors. In addition, some of the monitoring methods (e.g., use of emission factors) may not
require the installation of any monitoring equipment, and emission assessments may be done at
any time during the year, and measurements do not necessarily need to be undertaken during the
first quarter. Please see response to comment EPA-HQ-OAR-2009-0923-1011-27. However,
EPA considered that it may not be feasible for all reporting entities to fully comply with 2011
reporting requirements and therefore will allow the use of best available monitoring methods for
1209
some sources. Please see the preamble Section II.F. Regarding the comment on measurement
frequency, please see the response to EPA-HQ-OAR-2009-0923-1151-64.

It is not industry standard to determine GOR by individual wellhead. Yates requests the use of
BAMM for the first reporting year.
Response: EPA will allow use of best available monitoring methods for some sources. Please
see the preamble Section II.F. However, EPA disagrees that special provisions are necessary for
wellhead GOR, as todays final rule allows for reporters to estimate GOR using your available
data.

Meters are not designed to monitor intermittent flow. As liquids unloading has previously not
generally metered, it will be difficult to estimate the orifice size and/or meter capacity in order to
provide direct data to calculate this. As the process of installing meters for this purpose will be
experimental, PAW proposes that BAMM is used for this estimation for the first reporting year.
For smaller wells, calculation methodology 2 would be most appropriate, however the number of
hours the well was vented is difficult to monitor. PAW requests the use of BAMM for the first
reporting year as operators attempt to resolve this issue.
Response: EPA disagrees that use of best available monitoring methods will be necessary for all
well venting for liquids unloading because of issues with meters, because todays final rule
allows the alternative option of using reservoir and well parameters to perform an engineering
calculation. Please see the preamble Section II.F.

Organization:
Commenter: Michael Leonard
The time between the publishing of the final rule in the Federal Register and the beginning of the
reporting period will likely be inadequate to implement the data gathering requirements as stated
in the final rule due to the enormous data gathering requirements. We propose that best available
methods be permitted in the first quarter of the 2011 reporting year
1210

Commenter:
EPA seeks comment on the proposal not to allow use of best available monitoring methods for
engineering calculations and/or emission factors are used. (page 23)
Currently, methane leak detector cameras are not on all segments of natural gas pipelines. Thus,
it is recommended that the EPA allow reporters to use best available monitoring methods.
MidAmerican believes that 18 months is a reasonable amount of time to allow an individual
company to acquire the necessary equipment and train staff on its operation. Flexibility is
important in lowering the economic burden and allowing the use of best available monitoring
methods for all parameters while developing the reporting capabilities.
Please see the preamble Section II.F. The commenter has misinterpreted the rule, as it does not
require leak detection on natural gas pipelines.
This will require installation of flow meters for this requirement in many instances. PAW
requests the use of BAMM for the first reporting year.
Yates will need to install flow meters for this requirement. Therefore PAW requests the use of
1211
Furthermore, there are significant numbers of sources requiring direct measurement. As required
in 98.234(b) Monitoring and QA/QC requirements, All flow meters, composition analyzers,
and pressure gauges that are used to provide data for the GHG emissions calculations shall use
measurement methods, maintenance practices, and calibration methods, prior to the first
reporting year and in each subsequent reporting year Calibrating all meters, composition
analyzers and pressure gauges PRIOR to the reporting year will be nearly impossible given the
number of components to be calibrated. At the least, operators should have the ability to use best
available data for the first quarter of 2011. Furthermore, acid gas removal stacks will require the
installation of not one, but TWO meters that must meet these requirements. Three months is
simply not enough time to implement the requirements outlined in the rule as written.
PAW recommends that EPA provides BAMM opportunities for the first reporting year.
Response: EPA will allow the use of best available monitoring methods for certain sources.
EPA is seeking comment on the proposal not to allow the use of best available monitoring
methods for part or all of the first year of data collection. Red Cedar believes that EPA should
allow the use of best available monitoring methods for the first year of data collection due to the
short timeframe for compliance that is being provided. Specifically, EPA should allow best
available monitoring methods for parameters subject to direct measurement, such as compressor
rod packing vents, to allow companies sufficient time to install the proper metering devices
and/or hire a contractor to conduct the monitoring.
1212

Allow best available data for at least year one, and preferably for the years one and two, for
difficult to measure sources (i.e. the Year 3 sources from above) based on published emission
factors and other emission estimation methods from the API Compendium and developed by
Noble for its ongoing annual GHG inventory program. Noble prefers the previous options over
this approach because this exercise would result in year one facility GHG emissions that differ
from subsequent reporting years based solely on methods. This would lead to confusion when
comparing subsequent reporting year data or among outside reviewers of the information.
Please see the preamble Section II.F. EPA will not allow the use of any method under BAMM
because the usefulness of the collected data would be questionable given that it would be
obtained using inconsistent methods. Please see The Final Mandatory GHG Reporting Rule
(Final MRR), (40 CFR part 98) Section II.G.
Pre-amble, page 23, "EPA seeks comment on the proposal not to allow use of best available
monitoring methods for part or all of the first year of data collection. Further, if commenter's
[sic] recommend that EPA allow the use of best available monitoring methods for a designated
time period (e.g., three months), EPA seeks comments on whether requests for use of best
available monitoring methods should only be approved for parameters subject to direct
measurement, or also in cases where engineering calculations and/or emission factors are used."
- Many of the proposed emission estimation methodologies are far beyond common practice for
GHG emission inventory development from an oil and gas operations perspective. Therefore,
CAPP recommends that EPA allow best available monitoring methods (BAMM) for the first
year of the program to provide the time required to develop the data gathering and management,
sampling and analysis and reporting systems necessary to meet the requirements of the reporting
rule.
- CAPP believes that BAMM should be applied to all calculation methodologies including: direct
measurement, engineering calculations, and emission factor calculations.

1213
EP A should allow for the use of best available monitoring methods during the first reporting
year for all parameters, including those subject to direct measurement, engineering calculations,
and/or emission factors
Proposed Rule Effective Date / Best Available Monitoring Methods
"EPA seeks comment on the proposal not to allow use of best available monitoring methods for
engineering calculations and/or emission factors are used." (75 FR 18612)
Resolute Comments:
Resolute is concerned that the proposed effective date (January 1, 2011) of Proposed Subpart W
does not provide sufficient lead-time for full implementation (e.g., interpret the rule, determine
applicability, training personnel, setting up a compliance program, establishing recordkeeping
systems and databases, etc.) from the time EPA finalizes the rule, which may not occur until
sometime in late summer or fall of 2010. As a result, Resolute recommends delaying the
effective date until January 1, 2012.
If EPA insists on a January 1, 2011 effective date, Proposed Subpart W should allow for the use
of best available monitoring methods during the first reporting year to lessen the burden of
reporting upon the industry while still collecting reasonably available data on GHG emissions.
EPA should allow the use of best available monitoring methods for all reporting parameters,
including those involving direct measurement, engineering calculations, and/or the use of
emission factors.
Please see the preamble Section II.F. EPA disagrees, and requires reporting of calendar year
2011, please see the proposed rule April 2010 preamble Section II.A.
1214
Screening based on best available data:

In the alternative, until monitoring has been conducted at a facility, EPA should allow potentially
affected Subpart W sources to employ best available data at a facility or onshore production
reporting area level to estimate GHG emissions and determine applicability against the 25,000
metric tons threshold based on the estimated emissions. The best available data includes 2006
IPCC Guidelines, U.S. GHG Inventory, DOE 1605(b), The Climate Registry, California Climate
Action Reserve and corporate industry protocols developed by the American Petroleum Institute,
the Interstate Natural Gas Association of America, and the American Gas Association and
measurement data developed internally by the reporters.
Companies should be allowed to use these best available methods to estimate total emissions
from facilities or onshore production reporting areas for which monitoring at Subpart W facilities
has not yet been conducted, either because the rule is new or because the facilities have not yet
been estimated to exceed 25,000 metric tons based on best available methods. Companies should
be permitted to employ the estimated results to decide whether those facilities that have
emissions over 25,000 metric tons. Once a facilitys estimated emissions based on best available
methods exceed 25,000 metric tons, the first year monitoring methods would be carried out per
the final rule at these screened facilities and reports for the first year will be from these
facilities that have been screened to emit over 25,000 metric tons CO2e.
Response: EPA plans to develop a screening tool that may be used to assist companies in
determining their applicability. Please see the preamble Section II.F.
As currently proposed, Subpart W will require those facilities to which the rule may apply to
have their data collection methods in place by January 1, 2011. Unlike those industry sectors
currently subject to EPA's mandatory reporting rule under the original April 2009 proposal,
facilities covered by Subpart W (and, indeed, those facilities attempting to determine whether
Subpart W applies to them, as discussed above in Section C of these comments) are not
permitted to collect "best available" data for a period of time to allow sources a reasonable
amount of time to phase in the more burdensome monitoring requirements of the proposal.
Because many, if not most, facilities do not have any protocol currently in place to gather this
data, and because any relevant information that has been collected is likely not centralized,
IOGA-WV urges EPA to authorize facilities potentially subject to Subpart W to use industryestablished emission factors for the determination of applicability and for the first year's
reporting to allow time to implement a the complete monitoring and measurement program
required by Subpart W. This approach would allow EPA to begin its collection of data for 2011,
to alleviate what is likely to be a very high demand for needed services to meet monitoring
1215
requirements, and to address concerns that there is currently a paucity of companies and
technicians with expertise in Subpart W's proposed monitoring methods.
Please see the preamble Section II.F. EPA plans to develop a screening tool that may be used to
assist companies in determining their applicability. Please see the preamble Section II.F.
Commenter:
Source Type:
CBM produced water emissions
Regulatory reference for calculation/ monitoring requirements:
40 CFR 98.233(r)
Monitoring requirements/parameters:
1. For onshore petroleum and natural gas production facilities and onshore natural gas processing
facilities, concentration of GHGi, CH4 or CO2, in produced natural gas or feed natural gas; for
other facilities listed in 98.230 (b)(3) through (b)(8),GHGi equals one.
2. Total time the specific source associated with the fugitive emission was operational in the
reporting year, in hours.
Comment:
This source [CBM produced water emissions] is not addressed in The Climate Registrys
protocol for Oil and Natural Gas GHG reporting, which is largely a more-inclusive program than
EPA. If TCR did not address this as a source of emissions, it is possible that this source is not
expected to be a significant emitter of GHGs. As measuring emissions from this activity is
currently not industry standard, Yates requests the use of BAMM for the first reporting year.
Response: EPA has removed monitoring emissions from CBM produced water in todays final
rule. Please see the preamble Section II.E and response to EPA-HQ-OAR-2009-0923-1151-129
in todays final rule.
1216
VOLUME 13: SELECTION OF PROPOSED GHG EMISSIONS

CALCULATION AND MONITORING METHODS
13.0
SELECTION OF PROPOSED GHG EMISSIONS CALCULATION AND

MONITORING METHODS
Commenter:

Direct Measurement
Although EPA has reduced some direct measurement requirements from the first Subpart W
proposal, EPA still chooses to ignore longstanding and accepted international protocols for
emissions calculations that other countries have adopted and currently use to calculate their oil
and gas sector emissions. This includes Australia NGERS, EU-ETS, and Alberta for which all
emissions are based on the API Compendium-IPIECA standards. EPA is assuming that better
data will result from documenting and calculating emissions from every single small component
related to oil and gas production instead of using carefully documented emissions factors applied
across larger units. This procedure ignores the rules of probability and statistics and rejects
average values in favor of counting every bean. There is no reason to believe that this
burdensome method of data collection will produce an emissions number of greater value or
validity than other accepted and proven techniques. The proposed rule is considerably more
burdensome for onshore facilities than offshore facilities where MMS GOADS reporting
standards are acceptable.
Response: EPA evaluated existing monitoring methods and determined the most appropriate
methodologies to collect quality data for the purposes of this reporting rule while managing
burden to the industry. In several cases in this rule, EPA used monitoring methods that are
common practice in the industry and referenced in the API Compendium. For example, the use
of simulation software programs for dehydrators and tanks, mass balance approach for AGR
units, and engineering estimation of blowdown vent stack emissions.
In cases where EPA determined that existing emissions factors are not appropriate for the
purposes of this reporting rule, EPA used the most recent measurement data available to develop
new factors. For further details, please see rulemaking docket EPA-HQ-OAR-2009-0923 under
Revisions to Processing Leaker Emission Factors in Rule Table W-2. The API Compendium
provides emissions factors that are based on the Gas Research Institute study 182 that are known
to underestimate emissions from several key sources. For more information regarding these
182
EPA/GRI (1996) Methane Emissions from the Natural Gas Industry. Prepared by Harrison, M., T. Shires, J.
Wessels, and R. Cowgill, eds., Radian International LLC for National Risk Management Research Laboratory, Air
Pollution Prevention and Control Division, Research Triangle Park, NC. EPA-600/R-96-080a.
1217
sources please see the Technical Support Document (TSD) for todays final rule found in docket
(EPA-HQ-OAR-2009-0923). For these sources, EPA has either developed new emission factors,
required engineering calculations, or required the direct measurement of emissions while
considering the burden for each quantification methodology.
EPA agrees with the commenter on counting of components in onshore production. In todays
final rule, EPA has replaced individual component counts and population based emissions
factors for onshore petroleum and natural gas production with major equipment counts and
population emission factors. To further reduce burden, EPA has also provided default average
component counts for major equipment. For further clarification, please refer to Section II.D of
With regards to offshore operations, please see the response to EPA-HQ-OAR-2009-0923-1155

We suggest that the primary focus of the EPA's Subpart W rule should be to focus on improving
the quality in the inventory estimates through a reporting rule that:
encourages innovation in fugitive emissions monitoring and control measures,
expands industry experience and knowledge on fugitive emissions inventory monitoring and
reporting,
drives continuous improvement in the quality of the data, and
promotes cost-effective emissions reductions.
The proposed rule, as currently drafted, is very prescriptive in its use of specified emission
factors and monitoring methods, requiring all Local Distribution Companies (LDCs) to report
using the same factors and methods. We are concerned that this "consistency" will be
misinterpreted as "accuracy" by the public, stakeholders and regulators, when in fact a reporting
rule that permits alternative methods and use of more accurate data will provide better national
estimates and drive more informed policy and regulation.
In our view, there are policy approaches that will result in better outcomes. For example,
Canadian transmission and distribution companies are proposing to manage fugitive emissions
under their corporate asset management systems, much as they undertake pipeline integrity
management for safety purposes. This requires defined objectives and measured performance.
This asset management approach will be:
1218
adaptable and change over time as new technology or methods are adopted;
results oriented as performance measurement is required and operators who best know the
system determine how to meet objectives and are accountable for these; and
core to each company's business, by aligning environmental risk management (i.e. fugitive
emissions) with safety, integrity. security and other risks.
Response: The data reporting requirements for subpart W were designed to collect GHG
emissions data from petroleum and natural gas systems to inform future policy. Todays final
rule requires activity data be reported and measurement of emissions where there is very limited
emissions data publicly available. In regard to LDCs, it allows for the use of emissions factors
for smaller sources, and provides leaker factors in conjunction with leak detection where
emissions sources are large in magnitude. EPA disagrees with the assertion that subpart W is too
prescriptive. Todays final rule has the intent of informing future policy only and not requiring
any emissions reduction. EPA requires monitoring methods that are a balance between
consistency and accuracy of emissions that will be reported. Accurate reporting of emissions
from LDC would have required not only leak detection but also measurement of those emissions.
However, to mitigate cost concerns, EPA has limited leak detection and does not require direct
measurement. In addition, todays final rule provides methods for tracking emissions reductions
should the reporters choose to voluntarily perform them, such as doing multiple leak surveys and
applying leaker emission factors for the period that leaks are determined to occur. Also, as with
other rules in the past, EPA may revisit methodologies and allow for alternative work practices
or newer data to improve results as such data comes available and at the Agencys discretion.
Hence, EPA does not agree that an inconsistent characterization of emissions from the industry
can sufficiently inform future policy and requires uniform monitoring and reporting requirements
in todays final rule. In regards to the use of an asset management approach, EPA disagrees with
the comment to allow reporters to choose monitoring methods under an asset management
approach that are not specified in subpart W. Please see Section II.L of the preamble to The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) that provides further
details on consistency requirements across the rule. Finally, todays final rule does not address
emissions reduction, only emissions reporting.

We suggest that EPA consider modifying the proposed rule to reduce compliance costs. In
particular, we recommend that EPA remove the current requirement for mandatory site leak
detection at this time and allow use of population emission factors. An alternate option would be
to remove or decrease leak detection frequencies over time when the field data demonstrate that
reduced leak survey frequencies are warranted. The reporting rule should also allow LDC's the
1219
option to file based on alternative quantification methods or emission factor data where
improved data accuracy can be achieved.
Response: EPA is retaining the annual equipment leak detection requirements for natural gas
distribution, however EPA has provided more flexibility and reduced the compliance cost by
allowing facilities to use alternative leak detection equipment. For further clarification, please
refer to Sections II.E and II.F of the preamble to todays final rule.
With regard to the use of population factors, todays final rule allows for some above ground
metering and regulator stations to be monitored using station-level emission factors.
Pipe Data by Material Type: Not all LDCs will have complete data on mains and services broken
down by pipe material and the EPA rule should acknowledge this reality.
Response: EPA disagrees that not all LDCs will have complete data on main and services
broken down by pipe material. EPA has determined that the pipe material data already reported
to the Department of Transportation, per CFR title 49, part 191, Section 11, is sufficient for the
purposes of this reporting rule.
Page 18637-18645. 98.233. Calculating GHG emissions: It is unclear whether paragraphs (a)
through (r) and (u) through (z) apply to both onshore and offshore facilities or only to onshore
facilities. Since paragraph (s) specifically refers to offshore facilities, we assume that all other
paragraphs in this section refer only to onshore facilities. Please consider clarifying this and,
perhaps, reorganizing this part such that the offshore paragraph is at the beginning or end rather
than in the middle of the numerous onshore paragraphs.
Response: Please refer to Section 98.232 of todays final rule for a complete list of the sources
included under each industry segment; each industry segment need only report emissions sources
listed under the particular industry segment in Section 98.232. Section 98.233(s) of todays final
rule is a stand alone paragraph that outlines the GOADS reporting requirements for offshore
petroleum and natural gas producers. EPA has reviewed the content and placement of Section
98.233(s) in todays final rule and has determined that the organization and placement of that
section outlines the requirements for offshore operators.
1220

MMS prepares emission inventories for OCS facilities in the Gulf of Mexico every three years.
The last emissions inventory was prepared for the year 2008. The next inventory is planned for
the year 2011. However, since MMS calculates emissions for all facilities, which number in the
thousands, the complete inventory will not be available until sometime in 2013. We can assume
therefore that the facilities would need to perform their own calculations in order to meet the
March 31, 2012 reporting deadline. We can also assume that the facilities will be required to
report emissions annually. However, this is not clearly stated in the rule. Clarification is needed
for the final rule.
Response: EPA disagrees that reporters in the Gulf of Mexico OCS will need to calculate their
own emission for calendar year 2011. Please see Section II.D of the preamble to todays final
rule for how EPA intends to collect emissions data from offshore reporting in the first reporting
year.
Section 98.232(b) GHGs to report for offshore petroleum and natural gas production
There is a significant disadvantage in the reporting requirements for onshore versus offshore
operators, even though similar emission sources and practices are used. This difference is not
explained in the TSD or the proposed rule preamble.
For offshore oil and gas production, the GOADS program relies on counts of emissions sources,
volumes of throughputs from significant pieces of equipment, and parametric data to account for
non-combustion emissions. This is in sharp contrast to the requirements proposed for onshore oil
and gas production operations. In addition, since the Mineral Management Services only collects
GOADS data once every three years, the EPA proposes that offshore operators scale emissions
based on changes in production rates for the interim years.
Onshore operators should be allowed to use similar methodologies based on throughputs and
counts of major equipment. In addition, onshore oil and gas production operations should be
allowed to scale emission estimates for two out of three reporting years.
Response: In todays final rule, EPA has modified several reporting requirements to reduce the
burden for reporting onshore operators. Please see Section II.E of the preamble to todays final
rule for a detailed description of the changes made since the April 2010 proposal.
The commenter does not indicate which sources require more rigorous monitoring in the rule in
comparison to GOADS and hence EPA has not provided a prescriptive comparison of each
source. For further details, please see response to EPA-HQ-OAR-2009-0923-1155-14. Finally,
1221
EPA disagrees with the two to three year reporting cycle for onshore production. Onshore
production emissions are significantly higher than offshore emissions and hence EPA has
retained the yearly monitoring of onshore production emissions sources.
Proposed Rule Section 98.232(b) GHGs to report for Offshore Petroleum and Natural Gas
Production.Onshore petroleum and natural gas production operators should be allowed to use
the same approach as the offshore petroleum and natural gas production operators due to similar
emission sources, functional use and production practices.
Response: EPA disagrees with this comment. However, after considering several comments, in
todays final rule, EPA is now including alternative methodologies in some cases to reduce the
burden for onshore operators reporting. Please see Section II.E of the preamble to todays final
rule for a detailed description of the changes made since the April 2010 proposed rule found in
docket (EPA-HQ-OAR-2009-0923-0002).
Commenter: Kim Dang
Comments on the Proposed Quantification Methodologies. As stated above, Kinder Morgan
generally supports EPAs overall approach of requiring engineering estimates or emission factors
to be used to quantify emissions from the onshore petroleum and natural gas sector where
possible, and minimizing the use of direct measurement methods. Engineering estimates and
emission factors have a long track record in our industry, are well-understood and easy to apply
consistently, and are much more cost-effective than direct measurement. However, Kinder
Morgan makes the following suggestions and technical comments to EPAs proposed
quantification methodology for several source types.
1. Best Available Monitoring Methods. Kinder Morgan requests that EPA reconsider its decision
not to propose allowing BAMM to be used for all or part of the first year of compliance with the
proposed Subpart W. 183 As proposed, Subpart W would require Kinder Morgan to install a
considerable number of meters at acid gas removal units, centrifugal compressor wet seal
degassing vents, vents at reciprocating compressor rod packing cases, transmission storage tanks,
and potentially other locations. Installing these meters is costly and time-consuming process in
itself, and could result in substantial disruption to our operations since equipment must
frequently be taken off-line in order to install a meter. Component surveys will also have to be
conducted at facilities that require application of population emission factors, and do not have
183

1222
existing component counts. As explained below, Kinder Morgan also expects that the demand
for optical imaging instruments and the services of companies that conduct leak surveys and
measurements will exceed supply during at least the first year of compliance with Subpart W.
Even if service companies can meet the demand, Kinder Morgan is concerned that services will
be of low quality.
Kinder Morgan believes EPA has underestimated the amount of time that will be required to
carry out these tasks, and has proposed an optimistic implementation schedule for the proposed
Subpart W that leaves no room for error and creates substantial compliance risk. Our field testing
of the proposed Subpart W methodologies at one of our natural gas transmission compression
facilities indicates that leak detection and measurement at a single facility takes an experienced
two-person team at least one full day to conduct. Additional preparation and follow up work is
also required, causing the length of the inspection at a single facility to extend beyond two days.
Importantly, our field tests did not involve testing reciprocating compressor rod packing venting
in all three operating modes, as the proposed Subpart W would require. Three-mode testing of
these units would require reciprocating compressors to be forcibly cycled through operating,
operating depressurized, and standby modes, which would have significantly lengthened the time
required to carry out the emissions survey. To conduct leak detection as proposed, Kinder
Morgan and any company subject to Subpart W needs personnel, training, optical imaging
equipment, temporary and permanent flow meters, and measurement equipment such as but not
limited to calibrated bags and high volume samplers. In addition to purchasing equipment
personnel will need to be trained on how to use the equipment, the equipment will need to be
installed where applicable, and calibrated prior to January 1, 2011.
If the rule is finalized in September 2010, Kinder Morgan does not believe there will be
sufficient time left in the year to complete this work to assure compliance with Subpart W
beginning January 1, 2011. Kinder Morgan fully expects our field testing experience to apply to
the other industry segments included in Subpart W. Thus, Kinder Morgan urges EPA to consider
allowing BAAM to be used for the first year of reporting under the proposed rule. This approach
would allow EPA to collect data on oil and gas sector emissions during 2011, while allowing our
industry sufficient time to prepare for full implementation beginning 2012.
Response: In todays final rule, EPA is allowing Best Available Monitoring Methods for certain
sources and time periods. For more detailed information, please refer to Section II.F of the
preamble to todays final rule. EPA has also clarified several misinterpretations of the rule that
would have required substantially more effort from reporters than was intended by EPA. Please
see Section II.E and II.F of the preamble to todays final rule for examples of these
misinterpretations and how EPA has clarified them. Also, please see Section II.E of the
preamble to todays final rule for how EPA has simplified many of the monitoring requirements
to reduce the effort to monitor and report emissions. With these clarifications and simplifications
in monitoring methods and allowing BAMM for certain sources and time periods as noted above,
EPA concluded that reporters will have sufficient time to comply with the rule.
Commenter: Kim Dang
1223

Emissions Passing Through Manifold Vents. It is common for multiple emitting components in
petroleum and natural gas systems to be attached to a single manifold vent. In such cases, it is
most practical and accurate to monitor emissions at the manifold vent, rather than at each
individual component attached to the manifold. However, the proposed Subpart W does not
specify a monitoring method for manifold vents, and instead appears to require monitoring of
individual components regardless of the vent configuration. Kinder Morgan recommends that
EPA amend the proposed rule to allow for emissions monitoring at manifold vents where
appropriate.
9. Gas Driven Pneumatic Pump Venting. For purposes of calculating vented emissions from gasdriven pneumatic pumps, Kinder Morgan believes that engineering calculations are the most
efficient way to obtain the volume of liquid pumped annually by individual devices. However,
the proposed methodology states that a log of the liquid pumped annually must be
maintained. 184 Kinder Morgan requests that EPA clarify this methodology to state that
engineering calculations are an acceptable way to measure the volume of liquid pumped by
individual gas driven pneumatic pump.
Response: EPA is requiring emissions information to be reported by source type so that it can
better inform policy. When different emissions sources are connected to a common manifold the
reporter will provide a single estimate of emissions, which is not sufficient information for EPA
to determine where the emissions are occurring. Hence, in todays final rule, EPA has retained
the requirement for reporting of emissions from individual equipment.
For the comment on natural gas pneumatic pump venting, EPA now requires the use of
emissions factors for all pneumatic devices and pneumatic pumps in todays final rule. Hence,
the reporters only need to count the number of devices to estimate emissions. See Section II.E of
the preamble to todays final rule for further details.
Commenter: Kim Dang
SECTION 98.233 Calculating GHG emissions.
(c) Natural gas driven pneumatic pump venting. Calculate emissions from natural gas driven
pneumatic pump venting as follows:
(1) Calculate emissions using manufacturer data.
(i) Obtain from the manufacturer specific pump model natural gas emission (or manufacturer
184
Proposed 40 C.F.R. SECTION 98.233(c).

1224
gas consumption) per unit volume of liquid circulation rate at pump speeds and operating
pressures.
(ii) Maintain a log of Measure or estimate the amount of liquid pumped annually from
individual pumps.
(iii) Calculate the natural gas emissions for each pump using Equation W-3 of this section.
Es,n = Fs * V
Where:
(Eq. W-3)
Es, n = Annual natural gas emissions at standard conditions in cubic feet per year.
Fs = Natural gas driven pneumatic pump gas emission in emission per volume of liquid pumped
at operating pressure in scf/gallon at standard conditions, as provided by the manufacturer.

V = Volume of liquid pumped annually in gallons/year.
(iv) Both CH4 and CO2 volumetric and mass emissions shall be calculated from volumetric
natural gas emissions using calculations in paragraphs (u) and (v) of this section.
(2) If manufacturer data for a specific pump in Equation W-3 is not available, then use data for a
similar pump model, size and operational characteristics to estimate emissions.
(d) Acid gas removal (AGR) vent stacks. For AGR (including but not limited to processes such
as amine, membrane, molecular sieve or other absorbents and adsorbents), calculate emissions
for CO2 only (not CH4) using Equation W-4 of this section.
Ea, co2 = (V1 * %Vol1) (V2 * %Vol2) (V3 * %Vol3) (Eq W-4)
Where:
Ea, co2 = Annual volumetric CO2 emissions at ambient condition, in cubic feet per year.
V1 = Metered total annual volume of natural gas flow into AGR unit in cubic feet per year at
ambient conditions, as measured by meter or estimated using engineering methods.
%Vol1 = Volume weighted CO2 content of natural gas into the AGR unit.
V2 = Metered total annual volume of natural gas flow out of the AGR unit in cubic feet per year
at ambient conditions, as measured by meter or estimated using engineering methods.

%Vol2 = Volume weighted CO2 content of natural gas out of the AGR unit.
V3 = Total annual volume of CO2 flashed off from the AGR unit and recompressed into a
CO2 injection system to be injected into the ground, as measured by meter or estimated using
1225
engineering methods.
%Vol3 = Volume weighted CO2 content of gas stream flashed off and captured for re
injection.
(1) If a continuous gas analyzer is installed, then the continuous gas analyzer results must be
used. If continuous gas analyzer is not available, quarterly gas samples must be taken to
determine %Vol1 and %Vol2 according to methods set forth in SECTION 98.234(b).
(2) Calculate CO2 volumetric emissions at standard conditions using calculations in paragraph
(t) of this section.
(3) Mass CO2 emissions shall be calculated from volumetric CO2 emissions using calculations
in paragraphs (u) and (v) of this section.
(4) As an alternative to the method in Eq. W-4, CO2 emissions from AGR units may be
calculated using an industry standard software such as ProMax.
(f) Well venting for liquids unloadings.
(1) The emissions for well venting for liquids unloading shall be determined using either of the
calculation methodologies described in paragraph (f)(1) of this section. The same calculation
methodology must be used for the entire volume for the reporting year.
(i) Calculation Methodology 1. For each unique well tubing diameter and producing
horizon/formation combination in each gas producing field where gas wells are vented to the
atmosphere to expel liquids accumulated in the tubing, a recording flow meter shall be installed
on the vent line used to vent gas from the well (e.g., on the vent line off the wellhead separator or
atmospheric storage tank) according to methods set forth in SECTION 98.234(b). Calculate
emissions from well venting for liquids unloading using Equation W-6 of this section.
E a,n = T * FR
(Eq. W-6)
Where:
Ea, n = Annual natural gas emissions at ambient conditions in cubic feet.
T = Cumulative amount of time in hours of well venting during the year.
FR = Flow Rate in cubic feet per hour, under ambient conditions as required in paragraph
(f)(1)(i)(A), (f)(1)(i)(B) and (f)(1)(i)(C) of this section.
Calculate natural gas volumetric emissions at standard conditions using calculations in paragraph
(t) of this section. Both CH4 and CO2 volumetric and mass emissions shall be calculated from
volumetric natural gas emissions using calculations in paragraphs (u) and (v) of this section.
1226
(A) The average flow rate per minute of venting is calculated for each unique tubing diameter
and producing horizon/formation combination in each producing field.
(B) This factor is applied to all wells in the field that have the same tubing diameter and
producing horizon/formation combination, multiplied by the number of minutes of venting from
all wells of the same tubing diameter and producing horizon/formation combination in that field.
(C) A new emission factor is calculated every other year for each reporting field and horizon.
(ii) Calculation Methodology 2. Calculate emissions from each well venting for liquids
unloading using Equation W-7 of this section.
E s, n = [(0.37 X 10-3) * CD2 * WD * SP * V] + [SFR * HR]
(Eq. W-7)
Where:
Es, n = Annual natural gas emissions at standard conditions, in cubic feet/year.
0.3710-3 = [pi(3.14)/4]/[(14.7*144) psia converted to pounds per square feet]
CD = Casing diameter (inches).
WD = Well depth (feet).
SP = Shut-in pressure (psig).
V = Number of vents per year.
SFR = Sales flow rate of gas well in cubic feet per hour.
HR = Hours that the well was left open to the atmosphere during unloading.
(A) Both CH4 and CO2 volumetric and mass emissions shall be calculated from volumetric
(B) [Reserved]
*****************
(h) Gas well venting during conventional well completions and workovers. Calculate emissions
from each gas well venting during conventional well completions and workovers using Equation
W-9 of this section:
Ea,n = V * T
(Eq. W-9)
Where:
1227
Ea, n = Annual emissions in cubic feet at ambient conditions from gas well venting during
conventional well completions or workovers.
V = Daily gas production rate in cubic feet per minute.
T = Cumulative amount of time of well venting in minutes during the year.
(i1) Calculate natural gas volumetric emissions at standard conditions using calculations in
paragraph (t) of this section.
(ii2) Both CH4 and CO2 volumetric and mass emissions shall be calculated from volumetric
(iii i) Blowdown vent stacks. Calculate blowdown vent stack emissions as follows:
(1) Calculate the total volume (including, but not limited to, pipelines, compressor case or
cylinders, manifolds, suction and discharge bottles and vessels) between isolation valves.
(2) Retain logs of the number of blowdowns for each equipment type.
(3) Calculate the total annual venting emissions using Equation W-10 of this section:
Ea,n = N * Vv
(Eq. W-10)
Where:
Ea, n = Annual natural gas venting emissions at ambient conditions from blowdowns in cubic feet.
N = Number of blowdowns for the equipment in reporting year.
Vv = Total volume of blowdown equipment chambers (including, but not limited to, pipelines,
compressors and vessels) between isolation valves in cubic feet.

(4) Calculate natural gas volumetric emissions at standard conditions using calculations
in paragraph (t) of this section. vnaVNE*,_
(5) Calculate both CH4 and CO2 volumetric and mass emissions from volumetric natural gas
emissions using calculations in paragraphs (u) and (v) of this section.
(j) Onshore production and processing plant produced liquid storage tanks. For emissions from
atmospheric pressure storage tanks receiving produced liquids from onshore petroleum and
natural gas production facilities (including stationary liquid storage not owned or operated by the
reporter) and onshore natural gas processing plants facilities, calculate annual CH4 and CO2
emissions using the latest software package for E&P Tank (incorporated by reference, see
SECTION 98.7).
1228
(1) A minimum of the following parameters must be used to characterize emissions from
liquid transfer to atmospheric pressure produced liquid storage tanks.
(i)
Separator oil composition.
(ii)
Separator temperature.
(iii)
Separator pressure.
(iv)
Sales oil API gravity.
(v)
Sales oil production rate.
(vi)
Sales oil Reid vapor pressure.
(vii)
Ambient air temperature.
(viii)
Ambient air pressure.
********
(k) Transmission condensate storage tanks. For condensate storage tanks without vapor recovery
or thermal control devices in onshore natural gas transmission compression facilities calculate
annual emissions as follows:
(1) Monitor tank vapor vent stack for emissions using an optical gas imaging instrument
according to methods set forth in SECTION 98.234(a)(1) for a duration of 5 minutes.
(2) If the tank vapors are continuous and sufficient to indicate dump valve malfunction then use
a meter, calibrated bag, or high volume sampler to measure tank vapors as described in this
paragraph (2)(i) or utilize the E&P Tanks software program.
(i) Use a meter, such as, but not limited to a turbine meter, calibrated bag, or a high volume
sampler to estimate tank vapor volumes according to methods set forth in SECTION 98.234(b),
(c) or (d).
(ii) Use the appropriate gas composition in paragraph (u)(2)(iii) of this section.
(3) Calculate emissions from transmission condensate storage tanks to flares as follows:
(i) Use the storage tank emissions volume and gas composition as determined in paragraph (j)(1)
of this section.
(ii) Use the calculation methodology of flare stacks in paragraph (n) of this section to
determine storage tank emissions from the flare.
(l) Well testing venting and flaring. Calculate well testing venting and flaring emissions
as follows:
(1) Determine the gas to oil ratio (GOR) of the hydrocarbon production from each well
tested.
(i) If GOR is not available then use an appropriate standard method published by a
consensus-based standards organization to determine GOR.
1229
(ii) [Reserved]
*******
(4) Calculate GHG volumetric emissions at actual conditions using Equations W-13, W- 14, and
W-15 of this section.
Ea, i (un-combusted) = Va * (1-n) * Xi
(Eq. W-13)
Ea, CO2 (combusted) = (j) nj * Va * Yj * Rj
(Eq. W-14)
Ea, i (total) = Ea, i (combusted) + Ea, j (um-combusted)
(Eq. W-15)
Where:
Ea, i (un-combusted) = Contribution of annual uncombusted GHG i emissions from flare stack in
cubic feet, under ambient conditions.
Ea, CO2 (combusted) = Contribution of annual emissions from combustion from flare stack in
cubic feet, under ambient conditions
Ea, i (total) = Total annual emissions from flare stack in cubic feet, under ambient conditions
Va = Volume of natural gas sent to flare in cubic feet, during the year.
H = Percent of natural gas combusted by flare (default is 98 percent).
Xi = Concentration of GHG i in gas to the flare.
Yj = Concentration of natural gas hydrocarbon constituents j (such as methane, ethane, propane,
butane, and pentanes plus).
Rj = Number of carbon atoms in the natural gas hydrocarbon constituent j; 1 for methane, 2 for
ethane, 3 for propane, 4 for butane, and 5 for pentanes plus).
******
(p) Reciprocating compressor rod packing venting. Calculate annual CH4 and CO2 emissions
from each reciprocating compressor rod packing venting as follows
(1) Estimate annual emissions using a meter flow measurement using Equation W-17 of this
section.
Ea,i = MT * T * Mi
(Eq. W-17)
1230
Where:
Ea,i = Annual GHG i (either CH4 or CO2) volumetric emissions at ambient conditions.
MT = Meter volumetric reading of gas emissions per unit time, under ambient conditions.
T = Total time the compressor associated with the venting is operational in the reporting year.
Mi = Mole percent of GHG i in the vent gas; use the appropriate gas compositions in paragraph
(u)(2) of this section.
(2) If the rod packing case is connected to an open ended vent line then use one of the following
two methods to calculate emissions.
(i) Measure emissions from all vents (including emissions manifolded to common vents)
including rod packing, unit isolation valves, and blowdown valves using bagging according to
methods set forth in SECTION 98.234 (c).
(ii) Use a temporary meter such as, but not limited to, a vane anemometer or a permanent meter
such as, but not limited to, an orifice meter to measure emissions from all vents (including
emissions manifolded to a common vent)including rod packing vents, unit isolation valves, and
blowdown valves according to methods set forth in SECTION 98.234(b).
(3) If the rod packing case is not equipped with a vent line use the following method to estimate
emissions:
(i) You must use the methods described in SECTION 98.234 (a) to conduct annual leak detection
of fugitive emissions from the packing case into an open distance piece, or from the compressor
crank case breather cap or vent with a closed distance piece.
(ii) Measure emissions using a high flow sampler, or calibrated bag, or appropriate meter
according to methods set forth in SECTION 98.234(d).
(4) For each compressor record the operational mode that occurs during the measurement.
Conduct one measurement for each compressor in each of the operational modes that occurs
during a reporting period: (i) Operating. (ii) Standby pressurized. (iii) Not operating,
depressurized.]
(5) Calculate CH4 and CO2 volumetric emissions at standard conditions using calculations in
(6) Estimate CH4 and CO2 volumetric and mass emissions from volumetric natural gas emissions
using the calculations in paragraphs (u) and (v) of this section.
(q) Leak detection and leaker emission factors. You must use the appropriate standard methods
described in SECTION 98.234(a) to conduct an annual leak detection of fugitive emissions from
1231
all sources listed in SECTION 98.232(d)(9), (e)(7), (f)(5), (g)(3), (h)(4), and (i)(1). This
paragraph (q) applies to emissions sources in streams with gas content greater than 10 percent
CH4 plus CO2 by weight. Emissions sources in streams with gas content less than 10 percent CH4
plus CO2 by weight do not need to be reported. If fugitive emissions are detected for sources
listed in this paragraph, calculate emissions using Equation W-18 of this section for each source
with fugitive emissions.
Es,i= Count * EF * GHGi * T
(Eq. W-18)
Where:
Es,i = Annual total volumetric GHG emissions at standard conditions from each fugitive source.
Count = Total number of this type of emission source found to be leaking.
EF = Leaker emission factor for specific sources listed in Table W-2 through Table W-7 of this
subpart.
GHGi = For onshore natural gas processing plants facilities, concentration of GHG i, CH4 or
CO2, in the total hydrocarbon of the feed natural gas; for other facilities listed in SECTION
98.230(a)(3) through (a)(8), GHGi equals 1.
T = Total time the specific source associated with the fugitive emission was operational in the
(1) Calculate GHG mass emissions in carbon dioxide equivalent at standard conditions using
calculations in paragraph (v) of this section.
(2) Onshore natural gas processing plants facilities shall use the appropriate default leaker
emission factors listed in Table W-2 of this subpart for fugitive emissions detected from valves;
connectors; open ended lines; pressure relief valves; meters; and centrifugal compressor dry
seals.
(3) Onshore natural gas transmission compression facilities shall use the appropriate default
leaker emission factors listed in Table W-3 of this subpart for fugitive emissions detected from
connectors; block valves; control valves; compressor blowdown valves; pressure relief
*****
(r) Population count and emission factors. This paragraph applies to emissions sources listed in
SECTION 98.232(c)(2), (c)(9), (c)(15), (c)(21), (d)(8), (e)(6), (f)(4), (f)(5), (g)(3), (h)(4), (i)(2),
(i)(3) and (i)(4), on streams with gas content greater than 10 percent CH4 plus CO2 by weight.
Emissions sources in streams with gas content less than 10 percent CH4 plus CO2 by weight do
not need to be reported. Calculate emissions from all sources listed in this paragraph using
Equation W-19 of this section.
1232
Es,i= Count * EF * GHGi * T
(Eq. W-19)
Where:
Es,i = Annual total volumetric GHG emissions at standard conditions from each fugitive source.
Count = Total number of this type of emission source at the facility.
EF = Population emission factor for specific sources listed in Table W-1 through Table W-7 of
this subpart.
GHGi = for onshore petroleum and natural gas production facilities and onshore natural gas
processing plants facilities, concentration of GHG i, CH4 or CO2, in produced natural gas or feed
natural gas; for other facilities listed in SECTION 98.230 (b)(3) through (b)(8),GHGi equals 1.
T = Total time the specific source associated with the fugitive emission was operational in the
*****
(2) Calculate GHG volumetric emissions at standard conditions by converting ambient
temperature and pressure of GHG emissions to standard temperature and pressure using Equation
Es,i = (Ea,i * (460 + Ts) * Pa ) / ( (460 + Ta) * Ps )
(Eq. W-21)
Where:
Es,i = GHG i volumetric emissions at standard temperature and pressure (STP) conditions.
Ea,i = GHG i volumetric emissions at actual conditions.
Ts = Temperature at standard conditions. (oF).
Ta = Temperature at actual emission conditions. (oF).
Ps = Absolute pressure at standard conditions (inches of Hg). Pa = Absolute pressure at ambient
conditions (inches of Hg).

(u) GHG volumetric emissions. Calculate GHG volumetric emissions at standard conditions as
specified in paragraphs (u)(1) and (2) of this section.
(1) Estimate CH4 and CO2 emissions from natural gas emissions using Equation W-22 of this
section.
Es,i = Es,n * Mi
(Eq. W-22)
1233
Where:
Es,i = GHG i (either CH4 or CO2) volumetric emissions at standard conditions.
Es,n = Natural gas volumetric emissions at standard conditions.
Mi = Mole percent of GHG i in the natural gas.
(2) For Equation W-22 of this section, the mole percent, Mi, shall be the annual average mole
percent for each facility, as specified in paragraphs (u)(2)(i) through (vii) of this section.
(i) GHG mole percent in produced natural gas for onshore petroleum and natural gas production
facilities. If you have a continuous gas composition analyzer for produced natural gas, you must
use these values in calculating emissions. If you do not have a continuous gas composition
analyzer, then quarterly samples must be taken according to methods set forth in SECTION
98.234(b).
(ii) GHG mole percent in feed natural gas for all emissions sources upstream of the demethanizer
and GHG mole percent in facility specific residue gas to transmission pipeline systems for all
emissions sources downstream of the de-methanizer overhead for onshore natural gas processing
plants facilities. If you have a continuous gas composition analyzer on feed natural gas, you
must use these values in calculating emissions. If you do not have a continuous gas composition
analyzer, then quarterly samples must be taken according to methods set forth in SECTION
98.234(b).
(iii) GHG mole percent in transmission pipeline natural gas that passes through the facility for
onshore natural gas transmission compression facilities.
(iv) GHG mole percent in natural gas stored in underground natural gas storage facilities.
(v) GHG mole percent in natural gas stored in LNG storage facilities.
(vi) GHG mole percent in natural gas stored in LNG import and export facilities.
(vii) GHG mole percent in local distribution pipeline natural gas that passes through the facility
for natural gas distribution facilities.
(v) GHG mass emissions. Calculate GHG mass emissions in carbon dioxide equivalent at
standard conditions by converting the GHG volumetric emissions into mass emissions using
Equation W-23 of this section.
Masss,i = Es,i * Pi * GWP * 10-3
(Eq. W-23)
Where:
1234
Masss,i = GHG i (either CH4 or CO2) mass emissions at standard conditions in metric tons CO2e.
Es,i = GHG i (either CH4 or CO2) volumetric emissions at standard conditions, in cubic feet.
i = Density of GHG i, 0.053 kg/ft3 for CO2 and 0.0193 kg/ft3 for CH4.
GWP = Global warming potential, 1 for CO2 and 21 for CH4.
(w) EOR injection pump blowdown. Calculate pump blowdown emissions as follows:
(1) Calculate the total volume in cubic feet (including, but not limited to, pipelines, compressors
and vessels) between isolation valves.
(2) Retain logs of the number of blowdowns per reporting period.
(3) Calculate the total annual venting emissions using Equation W-24 of this section:
Mas c,i = N * Vv * Rc * GHGi *10-3 (Eq. W-24)
Where:
Mas c,i = Annual EOR injection gas venting emissions in metric tons at critical conditions c
from blowdowns.
Vv = Total volume in cubic feet of blowdown equipment chambers (including, but not limited to,
pipelines, compressors, manifolds and vessels) between isolation valves.
Rc = Density of critical phase EOR injection gas in kg/ft3. Use an appropriate standard method
published by a consensus-based standards organization to determine density of super critical

EOR injection gas.
GHGi = Mass fraction of GHGi in critical phase injection gas.
(x) Hydrocarbon liquids dissolved CO2. Calculate dissolved CO2 in hydrocarbon liquids as
follows:
(1) Determine the amount of CO2 retained in hydrocarbon liquids after flashing in tankage at
STP conditions. Quarterly samples must be taken according to methods set forth in SECTION
98.234(b) to determine retention of CO2 in hydrocarbon liquids immediately downstream of the
storage tank. Use the average of the quarterly analysis for the reporting period.
(2) Estimate emissions using Equation W-25 of this section.
Masss, CO2 = Shl * Vhl
(Eq. W-25)
1235
Where:
Masss, CO2 = Annual CO2 emissions from CO2 retained in hydrocarbon liquids beyond tankage,
in metric tons.
Shl = Amount of CO2 retained in hydrocarbon liquids in metric tons per barrel, under standard
conditions.
Vhl = Total volume of hydrocarbon liquids produced in barrels in the reporting year.
(y) Produced water dissolved CO2. Calculate dissolved CO2 in produced water as follows:
(1) Determine the amount of CO2 retained in produced water at STP conditions. Quarterly
samples must be taken according to methods set forth in SECTION 98.234(b) to determine
retention of CO2 in produced water immediately downstream of the separator where
hydrocarbon liquids and produced water are separated. Use the average of the quarterly analysis
for the reporting period.
(2) Estimate emissions using the Equation W-26 of this section.
Mass, CO2 = Spw * Vpw
(Eq. W-26)
Where:
Masss, CO2 = Annual CO2 emissions from CO2 retained in produced water beyond tankage, in
metric tons.
Spw = Amount of CO2 retained in produced water in metric tons per barrel, under standard
conditions.
Vpw = Total volume of produced water produced in barrels in the reporting year.
(3) EOR operations that route produced water from separation directly to re-injection into the
hydrocarbon reservoir in a closed loop system without any leakage to the atmosphere are exempt
from paragraph (y) of this section.]
(z) Portable equipment combustion emissions. Calculate emissions from portable equipment
using the Tier 1 methodology described in subpart C of this part (General Stationary Fuel
Combustion Sources).
(aa) Unsafe-to-monitor and difficult-to-monitor components. Calculate emissions from
unsafe-to-monitor and difficult-to-monitor components using an appropriate emission factor
or emission rate.
Response: EPA has reviewed the commenters proposed edits. In todays final rule EPA now
requires the use of emissions factors for all pneumatic devices and pneumatic pumps. Hence, the
1236
reporters only need to count the number of devices to estimate emissions. See Section II.E of the
preamble for further details.
After considering several comments, EPA has determined that there is sufficient flexibility in
terms of acceptable methods for calculating GHG emissions from AGR units and has revised the
monitoring method for AGR units in todays final to provide multiple options and eliminate the
need for new meters to measure throughput. Please see the response to EPA-HQ-OAR-2009
In regards to changes in well venting for liquids unloading calculations, please see the response
EPA agrees with the comment regarding the typographical error in the gas well venting and
blowdown vent stacks paragraph and the necessary changes have been made in the todays final
rule.
With regards to the title of the emissions source for storage tanks in onshore production, in
todays final rule EPA requires the monitoring of only hydrocarbon liquids for onshore
production storage tanks. The source is not required to be monitored for processing facilities in
todays final rule. This will avoid any confusion with monitoring of non-hydrocarbon tanks in
onshore production.
EPA does not agree with the commenter on the title for transmission storage tanks. The intent of
this source is to monitor any flow through emissions in these transmission tanks from leaking
dump valves and not the flashing from the liquids. Therefore, any liquid tank that stores either
water or hydrocarbon can have flow through emissions from dump valves. Hence, EPA has
retained the title of the source type.
EPA does not agree with the commenter on suggested alternatives to meters to quantify
emissions from transmission storage tanks. For further clarification, please see the response to
EPA-HQ-OAR-2009-0923-1024-30.
In todays final rule, EPA has modified the equations related to flare stacks emission calculation
to capture un-combusted CH4 emissions, un-combusted CO2 emissions and combusted CO2
emissions.
In todays final rule, EPA has agreed to allow compressor venting measurement in the as found
mode. For further clarification, please see response to comment EPA-HQ-OAR-2009-0923
0055-16.
EPA does not concur with the change in the facility definition title for processing facilities. EPA
has developed the definition for processing facilities to capture all relevant sources of emissions
from gas processing facilities, including residue gas compression equipment dedicated to the
processing facility. However, EPA has revised in todays final rule the processing facility
definition. For further clarification, please see Section II.F of the preamble to todays final rule.
EPA disagrees with the exclusion of EOR hydrocarbon liquids dissolved CO2 from the reporting
rule. For further clarification, please see the response to EPA-HQ-OAR-2009-0923-1011-21.
EPA agrees with the exclusions of produced water dissolved CO2. For further clarification, please
see the response to EPA-HQ-OAR-2009-0923-1151-129.
1237
EPA has provided an equipment threshold for onshore production and natural gas distribution
external combustion portable and stationary equipment. Please see Section II.E of the preamble
to todays final rule for further details.
EPA disagrees with the commenter on the use of emissions factors or emissions rates for unsafeto-monitor or difficult-to-monitor as EPA has provided alternative methods such that safety
concerns are eliminated. EPA has included sufficient provisions in todays final rule to account
for unsafe-to-monitor or difficult to-monitor equipment. Please see the response to EPA-HQ

Additionally, the API Compendium in particular has been accepted as an international
standard for estimating GHG emissions across the oil and gas industry. For example,
Australia NGERS, EU-ETS, and Alberta emission inventories are all based on the API
Compendium. EPA is assuming that a significant increase in overall accuracy will result by
requiring more complex and costly estimation of GHG emissions from even the smallest
sources within the natural gas gathering and processing industry. This assumption is made
while overlooking the EPAs own data that indicates the natural gas gathering and processing
industry emits only a very small fraction of the total US GHG emissions annually. In
addition, the currently proposed Subpart W ignores the standards of statistical averaging.
While Anadarko supports the full use of the API Compendium to calculate GHG emissions
for all sources covered by the proposed Subpart W, at a minimum we suggest the application
of the API Compendium in place of any direct measurement methods. For the gas gathering
and processing sector, direct measurement is generally proposed to be limited to compressor
wet seal vents and rod packing venting. The current proposal potentially requires direct
measurement of compressor rod packing vents from every cylinder of a compressor in cases
where the vent lines are not tied to a common line, which is a common practice in gathering
compression facilities and production facilities. The prescribed methods for these sources is
flawed by assuming a snapshot measurement applied across an entire year is more accurate
than an emission factor based on well documented and controlled studies.
Response: Todays final rule does not require the reporting of emissions from gas gathering and
boosting segment of the industry, please see Section II.F of the preamble to todays final rule.
However, EPA disagrees with the commenter on the use of API compendium. For further details,
Organization:
1238

(98.233) Numerous source types and respective calculation methods are given to determine the
amount of CO2 being released to the atmosphere at a wellsite. It can be assumed that the
majority of CO2 being produced at a wellsite will either flash/vent/leak to the atmosphere at STP
conditions and the remaining dissolved CO2 in produced water and hydrocarbon liquids will be
negligible, due to STP conditions. The minimal quantity of dissolved CO2 does not warrant the
quarterly sampling burden. We propose that a single annual sample be taken that is
representative of the CO2 concentration at the wellhead. This concentration and the annual flow
of the well can be used to determine the mass of CO2 being produced at the well. This mass of
CO2 will flash to the atmosphere at STP conditions on sites where the products are not piped offsite under pressure. On sites that pipe products off-site under pressure, a second sample may be
needed to determine the concentration of the CO2 in the methane stream that would be
representative of the CO2 being transported offsite. A mass balance equation could then be used
to determine CO2 being flashed/vented/leaked to the atmosphere at the site. The second sample,
representative of what is transported off-site under pressure, would only be necessary if a
downstream processing facility was capturing a CO2 stream for the purpose of supplying CO2 to
a commercial/industrial customer. Otherwise, the CO2 being transported off-site under pressure
can also be assumed to go to atmosphere at a downstream location and therefore, a mass of CO2
being produced at the wellsite would best represent the quantity of CO2 ending up in the
atmosphere.
Response: EPA did not intend for the sampling of hydrocarbon liquids and produced water other
than at EOR sites that use CO2 injection. With regard to the annual sampling of hydrocarbon
liquids, please see the response to EPA-HQ-OAR-2009-0923-0582-40.
Todays final rule does not require the monitoring of CO2 from produced water at any onshore
petroleum and natural gas production operations. With regard to sampling of produced water,

Organization:
(98.230) It has been stated in the preamble to this subpart that mass balance equations are not
acceptable to calculate overall emissions due to the inaccuracy of the flow meters necessary to
determine flow rates. However, the concept of tolerance stacking may come into play when
considering multiple source categories and associated calculations. Tolerance stacking is well
known in several industries. In this scenario, multiple smaller systems each have an associated
acceptable calculation error. Though these calculation errors may be small when viewed
individually, they have the ability to upon themselves as additional systems with their own
calculation errors are added. A scenario can occur where the errors favor a high or low tolerance,
and therefore have the ability to add up. This may produce a significantly higher error than a
simple mass balance equation. We propose that a study be conducted to address the concept of
1239
error tolerance stacking and how it relates to the numerous source categories being aggregated
under the proposed Subpart W such that the most accurate overall calculation method is used.
Response: EPA disagrees with the commenter. EPA made the cited statement specifically in
relation to transportation pipelines. It is not clear from the comment how tolerance stacking can
be practically used to separate the error component from actual meter reading of gas throughput
in gas transmission systems. Hence, EPA does not have sufficient information from the
commenter to provide any further response.
Aka recommends that EPA remove the requirement in Section 98.236(b) for operators to report
emissions separately for standby equipment. First, the term standby is not defined anywhere
in the rule (neither Subpart W nor Subpart A). Moreover, equipment is generally not specifically
designated by operators as primary or standby. This adds an additional layer of reporting
that is unwarranted and confusing, uses nebulous terminology, and provides no meaningful
information. `
Response: After reviewing applicable comments from the industry, EPA has eliminated the
requirement to report emissions separately from standby equipment, except compressors, in
todays final rule. The emissions from isolation valves when a compressor is in standby
depressurized mode can be significant, which is why EPA requires these sources to be
monitored. Hence, EPA has removed the provision for reporting standby equipment in Section
98.236(b) and has simplified the requirements for monitoring of compressors in standby mode.
Please see the response to EPA-HQ-OAR-2009-0923-0055-16 for further details.
The rule should not require actual fugitive component count
With the addition of the onshore petroleum and natural gas production segment of the Subpart,
the scope of this Subpart has been vastly expanded, especially the inclusion of population count
of fugitive components. It would be extremely time consuming and cost intensive for individuals
to go out to approximately 1,800 tank batteries and over 6,000 wellheads in the Permian Basin
alone to count these components and document these data based on this Subpart's definition of
"facility". This would be a monumental task to conduct and maintain in every basin where
Pioneer operates in the U.S. Further, no laboratory has the infrastructure to handle the anticipated
volume of samples from Pioneer, as well as other operators combined. Moreover, an actual count
is not needed to develop a reasonable estimate of fugitive emissions. A population average from
similar batteries should provide reasonable results. Perhaps a representative battery could be
chosen in each "reporting unit" (per the reporting unit" recommendation in point 1).
1240
Alternatively, batteries could be grouped in each basin by their characteristics or size, based on
throughput, and a population average of components could be estimated for each size grouping.
Response: In todays final rule, EPA requires the monitoring of major equipment and not
individual components. For further details, please see Section II.D of the preamble to todays
final rule.
EPA disagrees with the commenter on the grouping of fields within a basin. For further details,
Section 98.236: Data Reporting Requirements
The following paragraphs in this section of the proposed rule request data reporting by basin and
field. If reporting is done at a basin level for comparison to the applicability threshold, the data
reporting requirements should be consistent with the basin-wide rollup of emissions and the data
should not have to be reported at a field level due to the additional burden that would be imposed
in order to delineate the data by field and the redundancy in reporting the field level data. BP
requests that the term field be deleted from the following paragraphs in this section: Section
98.236 (c)(6), (7), (8), (12), (13), (15), (16), (21), and (22)
Response: EPA disagrees with the commenter. Please see response to EPA-HQ-OAR-2009
Alternative Option for Reporting Onshore Petroleum and Natural Gas Production
EPA evaluated and is taking comment on one alternative option for reporting from onshore
petroleum and natural gas production; field level.
BP does not believe the regulatory named field level (as defined in the referenced EIA field list)
is a useful concept for either threshold or emission estimation purposes. With over 44,000 named
fields in the EIA list, the widely varying number of wells in each named field (from zero to tens
of thousands), and the dynamic nature of the list, it would introduce a significant amount of
uncertainty and additional burden. Of particular concern is the changeable nature of the list. As
State Oil and Gas agencies modify their field lists, the EIA subsequently modifies their list,
usually with a delay of about one year. For example, Wyoming is actively reviewing their named
fields with a goal of combining like fields (similar to the sub-basin proposal) and reducing the
1241
number of fields listed. This flux in field names and boundaries would pose a significant
challenge to an owner/operator in determining what wells and facilities to group for threshold
and other purposes under the rule.
C. Basin Versus Field Level Reporting
EPA seeks comments on our decision to propose the basin level approach, and whether there
would be advantages to requiring reporting at the field level instead.
As described above, BP proposes an alternative method of grouping fields within a basin, based
on common production characteristics. These groupings would be established and documented
by the reporting operator in the monitoring plan. Emissions from the groupings would be
summed at the basin level for comparison against the 25,000 tonne CO2e reporting threshold.
Response: EPA disagrees with the commenter on the EIA field code master. EPA has analyzed
the field code master for the last 4 years and found that the changes in the database are
insignificant. In fact, there were only 30 changes in field definitions between 2007 and 2008 of
the total 64,454 fields in the database. Similar numbers result from comparing 2006 with 2007
(170 changes in field definition of a total 63,873 fields in the database) and comparing 2006 with
2005 (44 changes in field definition of a total 63,356 fields in the database). Hence, EPA does
not consider these changes to be of any significantly dynamic nature. Therefore, although EPA
has not required the use of field code master as a facility definition, EPA has required the use of
the definition in monitoring methods such as for gas well venting during completions and
workovers with hydraulic fracturing. Also, EPA does not agree with the concept of grouping
fields within a basin; please see response to EPA-HQ-OAR-2009-0923-1305-46 for further
details.
Alternative Software Packages for Calculations
EPA seeks comment on whether there are additional or alternative software packages to E&P
Tank and GlyCalc that should be required to be used to calculate emissions.
The rule should allow reporters to use E&P Tanks and GlyCalc or any commercial process
simulator software (such as HySys or ProSim) to estimate emissions from tanks and glycol
dehydration. The rule should also not specify the version numbers of the software packages, as
these programs are updated periodically.
Response: EPA agrees with the commenter. With regard to software flexibility, please see the
1242

Specific Source Type Comments and Alternative Emission Estimation Methodologies
Following, BP proposes alternative estimation methodologies for many of the source types
applicable to the Onshore Petroleum and Gas Production Sector (OGPG) and other source
categories as applicable. These alternative methodologies are warranted to reduce the man power
and cost burden and to address errors in EPAs assessment of specific sources and
methodologies. Also, BP recommends that some of the source types be deleted from the final
rule because of the insignificant GHG emissions associated with them (reasons/examples are
provided below for relevant source type discussions). BP requests that the following 3 categories
be deleted: (1) Hydrocarbon Liquids Dissolved CO2, (2) Produced Water Dissolved CO2, (3) ?
Response: EPA does not agree with the exclusion of hydrocarbon liquids dissolved CO2 from
reporting. However, EPA has clarified that the source is limited to CO2 EOR operations only. For
further clarification, please see the response to EPA-HQ-OAR-2009-0923-1011-21.
EPA agrees with the exclusion of produced water dissolved CO2 from the reporting rule. With
regard to sampling of produced water, please see the response to EPA-HQ-OAR-2009-0923
1151-129.

Section 98.233(q) and (r) Fugitive Emissions
Onshore Production Source Category: Fugitive emissions from valves, connectors, open ended
lines, pressure relief valves, compressor starter gas vents, pumps, flanges, and other fugitive
sources (such as instruments, loading arms, pressure relief valves, stuffing boxes, compressor
seals, dump lever arms, and breather caps for crude services).
BP supports EPAs decision to base fugitive emission estimates for onshore production on
population count and emission factors. However, the methodology for arriving at a component
count must be significantly changed to enable a reasonable approach to arrive at a component
count for the hundreds of thousands of sites subject to the proposed rule. As currently structured,
the rule would require a physical component count (along with concurrently gathering other
information) at each individual well-site and associated facility. It is very likely that this amount
of work could not be done by the end of 2011 and the projected costs are very high. To address
this problem, BP suggests the following:
In a particular producing area (reporting unit or reporting area) operators tend to use similar
equipment packages, such as separators, compressors, dehydrators, and tanks, and similar site
1243
layouts or configurations. BP suggests that EPA modify the methodology for arriving at a
component count for a particular reporting unit as follows:
1. Require a component count for a small number (10 - 20) of each type of major equipment
package. Inventory the number of each type of major equipment packages deployed in the
reporting unit/area. Multiply the component counts by the number of major equipment packages
to arrive at a total component count for the major equipment packages.
2. Require a component count for a small number (10 - 20) of each type of site (e.g. single wellsite, multi-well site, tank battery, central production site, etc) for components which are not part
of the major equipment packages. Multiply the site component counts by the number of sites of
each particular type to arrive at a total component count for components not part of the major
equipment packages.
3. Sum the major equipment and site component counts for each reporting area to arrive at the
total component count for the reporting area and then apply the methodology specified in the rule
(Section 98.233 (r)) to arrive at a fugitive emission estimate for the reporting unit/area.
4. Sum the fugitive emission estimates for all reporting units/areas within an identified basin for
purposes of annual reporting.
By making these modifications to estimate component counts, EPA can achieve a more
appropriate balance between the burden of the rule and the amount of emissions covered while
still yielding good quality emission estimates. Adopting this methodology would also partially
address the issue of the industrys ability to physically meet the rule requirements in the time
contemplated by EPA.
A minimum component size for counting of components or leak surveys in the processing sector
should be defined (i.e., line size). BP suggests that components associated with line or tubing
less than or equal to 3/4-inch outside diameter should be excluded from both counts and leak
surveys. This is consistent with other regulations for leak detection and repair programs. Rather
than attempting to classify fugitive components by the type of fluid contained as depicted in
Tables W-1 and W-2, EPA should allow operators to simply assume that all components are in
gas service. If EPA does not enable this approach, then the rule should clarify the handling of
multiphase flow which is common in the oil and gas production sector.
Response: In todays final rule, EPA requires the monitoring of major equipment and not
individual components for onshore petroleum and natural gas production. For further details,
please see Section II.D of the preamble to todays final rule. Hence, the commenters suggested
methodology is no longer relevant.
In todays final rule, EPA does not require the monitoring of tubing systems equal to or less than
one half inch diameter. Also, EPA requires the counting of major equipment to estimate
emissions, as opposed to counting components; please see Section II.F of the preamble to todays
final rule for further details. EPA has clarified what major equipment belongs to oil or gas
1244
service in Section 98.233(r) of todays final rule and hence the issue on multiphase flow is no
more relevant.
Organization: Heath Consultants
Commenter: Milton W. Heath
Standardized Components (Flanges, Unions, Thread/Pipefittings, tubing connectors, valve
stems, ball-valves, grease fittings, valve caps, unloader flanges, actuators etc.)
The largest number of components at any gas facility will be in the area of what we categorize as
standardized components. At most facilities, you should expect to see a minimum of 2000
components and as many as 10,000. Screening these components has typically been done in the
past with the use of liquid bubble solution, ultrasonic leak detectors, leak screening with a pump
driven gas detector or flame ionization detector (OVA/TVA)
The procedure for an electronic gas leak detector that may use infrared, catalytic oxidation,
thermal conductivity or flame ionization, is to inspect the parameter of a connecting component
with a specialized survey probe whereby a sample is pulled into the device and analyzed for the
presence of hydrocarbons at a very low rate (PPM or % GAS). This process is labor intensive
and time consuming because the operator must inspect every component that has natural gas
pressure behind it which could possibly leak due to wear and tear, exposure to heat or vibration
etc.
Newer Remote Infrared Laser Leak Detectors/Screeners
Recently, in the last four years, the introduction of remote infrared lasers has allowed operators
to improve this screening process for standardized components. Now it is possible to screen
each connector at a distance of six feet to 100 feet away from the target source with the same
accuracy and the sensitivity of flame ionization (down to 10 Parts Per Million). One of the great
advantages of infrared laser leak detection is that the resulting indication of a leak at a given
component is repeatable and consistent with each operator, regardless of what their level of
experience or expertise is. So long as the laser spotter is used to show where the infrared laser is
screening on the component and leaking gas is present, an audible alarm will be instantaneously
heard while the concentration of that leak is displayed in parts per million metered on the
display. These two bits of information are not adjusted or manipulated as with an imaging
camera to insure a quality image, they are derived automatically and instantly because of the
superior quality of the infrared laser coupled with sophisticated micro-processor controls which
take the interpretation and mental fatigue away from the operator and put it into the screening
device where it belongs.
The procedure used for leak screening with a remote infrared laser detector is to:
1. Turn the power button on and allow one minute for the device to warm up
2. Calibrate the laser in the carrying case with a known quantity of gas safely contained in a
transparent hermetically sealed calibration cell (approximately 1% gas, non-explosive). This
takes about 1 minute.
3. Place the control unit over your shoulder using the shoulder carrying strap and hold the grip of
the transiever with your preferred hand and begin leak screening your target source.
1245
4. Working from one end of the piping down to the other, slowly screen the openwings of
flanges, valve stems, connectors by sweeping the laser spotter (built into the grip and depressed
with your index finger for controlled spotting when you need it) across the target source in a
zigzag motion from top to bottom. You can rapidly scan these areas to identify gas leaks which
can then be confirmed with a liquid bubble solution and tagged for future quantification or direct
measurement (based on your protocol).
IV. Leak Screening of Standardized Components with the Gas Imaging Camera versus
infrared laser technology.
The current proposed EPA rule for subpart W requires leak screening of standard components
such as the ones described above with a gas imaging camera and nothing else. This mandate is
difficult to comprehend, and an onerous one at that, given the enormous advantages of leak
screening with both conventional technology and even newer better technologies also described
above.
The fallacy of this mandate is that the gas imaging camera is an effective tool for this application
when in fact it is just the opposite. The time it takes to effectively screen a thousand or more
components is easily twice the time and as much as four times longer when compared with an
infrared active laser leak detector. Perhaps a better use of the camera would be to stick with the
vented components since the infrared laser must have a proper reflective background in order to
screen a component successfully. Application of these technologies is critical to insuring the
proper use and success of both these devices. Limiting the operator to one screening device is to
limit the effectiveness of the survey which in this case will yield poor results while increasing the
time and cost of the proposed survey.
Response: EPA has evaluated alternative methods for detection of equipment leaks for their
viability and comparative accuracy to the optical gas imaging instrument in the proposed rule. In
todays rule, EPA requires leak detection using one of several methods available, besides the
option of an optical gas imaging instrument. EPA is still requiring that inaccessible sources use
optical gas imaging instruments due to potential safety and cost concerns related to leak
detection of sources that cannot be physically accessed from a fixed, supportive surface with a
hand held leak detection device such as OVA/TVA, or which do not have a reflective
background for an IR laser detection device. For further details, please see Section II.F of the
preamble to todays final rule.
13.1
EMISSION FACTORS

1246
- The flare N2O emission factor in Table W-8 is based on MMCf gas production. An emission
factor based on volume of gas combusted would be more appropriate.
Response: In the final rule, EPA has replaced the proposed rules flare stack N2O emission
factor with a more accurate methodology to quantify N2O emissions; please see the response to
EPA-HQ-OAR-2009-0923-1027-11.
The emissions factors that EPA requires to be used in calculating GHG emissions from natural
gas systems were developed more than a decade ago by the Gas Research Institute (GRI). GRI
used 10 year old data at that time and the data was only from a limited amount of pipeline
components. These emissions factors were used to determine a ballpark estimate of nationwide
GHG emissions from natural gas systems, and the intent was not to use them in accurately
estimating individual company or source specific emissions that could potentially be used in
future compliance with climate change legislation or regulation.
Response: EPA disagrees that emission factors were solely developed from the GRI reports.
EPAs use of the GRI reports was governed by the lack of better publicly available data. In fact,
both the GRI studies and Canadian studies were used to develop emission factors for natural gas
distribution facilities. EPA has documented the data in the Canadian studies and the derivation
of emission factors in the rulemaking docket (EPA-HQ-OAR-2009-0923) under Revisions to
Processing Leaker Emission Factors in Rule Table W-2. and the Greenhouse Gas Emissions
from the Petroleum and Natural Gas Industry: Technical Support Document (TSD) found in
docket (EPA-HQ-OAR-2009-0923-0027)..
Accuracy of Emission Factors
NGA questions the accuracy and validity of the proposed emission factors and their application.
The proposed method of applying these emission factors to determine applicability or emission
rates in our view will require clarification.
We would also recommend that the proposed emission factors be peer-reviewed to determine
applicability, particularly to the LDC system.
Like AGA, we are concerned that EPA is planning to use emission factors that were developed
over a decade ago for another purpose - and that were based on limited field work conducted
nearly 20 years ago. Changes and upgrades on the distribution system are not reflected within
1247
these numbers and raise concerns about validity.

Finally, flexibility should be built into the rule to provide for updates when new emission factors
become available.
Response: EPA disagrees that the emission factors in the rule are not accurate or valid. The
commenter has not provided sufficient details on what clarification EPA should provide on the
application of emissions factors. EPA used the best publicly available data to develop the
emission factors for natural gas distribution systems in todays final rule. EPA has documented
its derivation of emission factors used in the rulemaking docket (EPA-HQ-OAR-2009-0923)
under Revisions to Processing Leaker Emission Factors in Rule Table W-2. EPA deemed the
use of leaker emission factors, and the associated data quality, sufficient to inform future policy.
If new and improved emission factors become available, at an appropriate time, EPA may re
propose applicable parts of subpart W. In addition, EPA will evaluate the natural gas
distribution data received through the MRR, which are based on emission factors and
methodologies in subpart W. EPA will also evaluate other emission factors that EPA may
receive. EPA may also consider using direct measurement of equipment leaks and vents if EPA
finds that the natural gas distribution emissions data received through the MRR is insufficient to
inform policy.
Finally, and ideally, EPA should postpone imposing Subpart W on LDCs until EPA and industry
partners have completed our ongoing joint research & development work to update the existing
emission factors. This work has been underway and has already produced updated information.
The remaining work could take approximately three to four years to complete, depending on the
resources available. It would be ironic and unfortunate if industry and agency resources had to be
diverted from developing more accurate emission factors in order to cope with the costs and
burdens of applying the old emission factors under Subpart W. As EPA recognizes, the existing
emission factors for our sector are outdated. They were developed a decade ago based on testing
facilities and equipment nearly 20 years ago. Distribution systems are different now, and
generally much tighter than they were 20 years ago as a result of improvements in both materials
and operational procedures. As a result, the existing emission factors tend to overstate the actual
emissions from distribution systems. The best way to obtain the information EPA seeks is to
complete the work to develop updated, more accurate emission factors.
Response: EPA disagrees with the assertion that reporting emissions from natural gas
distribution facilities should be delayed until updated emission factors become available. The
commenter has alluded to studies intended to improve emission factors for the LDCs. EPA
recognizes that these studies are being conducted to improve emission factors. However, EPA is
not willing to forfeit valuable data, intended to inform future policy. EPA will not postpone the
finalization of subpart W in anticipation of the results of these studies. If new and improved
emission factors become available, at an appropriate time chosen by EPA, EPA may re-propose
1248
applicable parts of subpart W. For this final rule though, EPA used the best publicly available
data to develop emission factors for the natural gas distribution segment. Please see the response
to EPA-HQ-OAR-2009-0923-1299-5.
.

Cost of Improving Emission Factors
It is well known that the accuracy of existing emission factors varies by as much as 300% to
2,000%. There have been significant efforts made to improve those accuracies within the last two
years by the American Gas Association and individual companies, such as Southwest.
Southwest, in collaborating with the Gas Technology Institute (GTI), has conducted direct
measurement programs to further enhance the base information for emission factors. The cost of
this work and the time needed to do the work does not appear to be adequately addressed in the
justification for the proposed rule.
EPA found that emission factors generated from the Clearstone studies related better to methanerich stream fugitives and were more appropriate than other emission factors developed for highly
regulated refinery and petrochemical plants on VOC emissions. Therefore, EPA is using
emissions data from the Clearstone studies as the basis for the leaker factors proposed in this
rule. Due to the extremely short comment period, Southwest does not have time to thoroughly
evaluate the emission factors from the Clearstone studies. Southwest has no comfort that this
alternative approach better estimates annual facility emissions without resulting in additional
reporting burden to the facilities.
Response: EPA solicited data in its proposed rule but the commenter has not submitted different
or improved emission factors. Furthermore, in developing emission factors for use in todays
final rule, EPA conducted a thorough search of all publicly available documents to determine the
best data to use to develop emission factors. EPA did not find the emission factors that the
commenter is alluding to. In addition, EPA does not intend to use emission factors that are not
publicly available. Consequently, EPA used the best available, public data to develop the
emission factors for natural gas distribution facilities in todays final rule. Please see the
response to EPA-HQ-OAR-2009-0923-1299-5 for further details. EPA does not need to include
in the rules cost burden analysis the cost incurred by reporters to conduct their own emission
measurement studies.

1249
In addition to the challenges described above, postponing implementation of Subpart W would

also allow the development of updated emission factors for distribution system equipment. This
emission factor update is currently underway by AGA, EPA, and the Gas Technology Institute
(GTI) and is expected to produce more accurate methane emission factors than the emission
factors developed over a decade ago by the Gas Research Institute (GTIs predecessor) using
data from testing a limited sample of equipment. The updated emission factors will be used to
facilitate more accurate GHG emission estimates for plastic distribution pipe, metering and
regulator (M&R) stations and other natural gas distribution equipment. This work is not expected
to be complete until 2012 or 2013 depending on funding availability, but it has already produced
a new emission factor for plastic distribution pipe that better reflects the extremely low leak rate
for modern plastic pipe than the old, statistically questionable emission factor EPA is currently
using to estimate emissions from natural gas distribution. Upon completion of this work, we
expect combined GHG emissions from natural gas distribution systems across the country to be
even lower than the already low estimate of less than 1% of total U.S. GHG emissions.
Response: The emission factors that the commenter cited were not submitted to EPA. Please see
the response to EPA-HQ-OAR-2009-0923-1020-4 for further details.

More accurate emissions factors are expected to be available in about 2012 to 2013 from studies
that are currently in progress, involving the collaboration of AGA, GRI and EPA. MDU
recommends that the EPA review the new emissions factors before requiring reporting of GHG
fugitive emissions from natural gas systems with the outdated emissions factors and
inadvertently creating long term erroneous climate policy determinations. The new emissions
data being developed from the new study is demonstrating in some cases that the old emissions
factors are overestimating emissions, such as in plastic pipe.
Response: EPA used the best publicly available data to develop emission factors for natural gas
distribution facilities. Please see the response to EPA-HQ-OAR-2009-0923-0049-11. If new and
improved emission factors become available, at an appropriate time, EPA may re-propose
applicable parts of subpart W. Please see response to comment EPA-HQ-OAR-2009-0923
1299-5.

Commenter:
EPA requests comments on the use of emission factors from the Clearstone studies. For further
1250
details see Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background TSD (EPA-HQ-OAR-2009-0923).
MidAmerican believes that studies from the Interstate National Gas American Association
(INGAA) have the most accurate and appropriate emission factors for natural gas transmission
systems. MidAmerican recommends, as an option, that the EPA use the emission factors
developed by INGAA.
Response: EPA disagrees with the commenters suggestion to use emission factors developed by
INGAA. The data used by INGAA to develop leaker emission factors is the same as the data
used by EPA. However, INGAAs emission factors are directly from the report Handbook for
Estimating Methane Emissions from Canadian Natural Gas System written by Clearstone
Engineering Ltd. for the Canadian Energy Partnership for Environmental Innovation. EPA took
this same data, in addition to other data collected by Clearstone Engineering Ltd., and developed
emission factors using EPAs own methodology. EPAs emission factors are based on a larger
data set than INGAAs emission factors. In addition, EPA was also able to segregate the data to
develop compressor related and non-compressor related emission factors to more accurately
quantify equipment leaks from onshore natural gas transmission compression. Further
information on the development of leaker emission factors for onshore natural gas transmission
compression can be found in the rulemaking docket (EPA-HQ-OAR-2009-0923) under
Revisions to Processing Leaker Emission Factors in Rule Table W-2

Emission Factors - The emission factors EPA uses in Subpart W are outdated and therefore tend
to overestimate actual emissions. EPA should postpone finalizing Subpart W until the emission
factors are updated. If that is not possible the current research and development to update these
emission factors should be a priority to complete in order to get the most accurate GHG
inventory. In the event that updated factors are found to be lower, they should be applied
retroactively to previous year's submissions.
Response: EPA disagrees with the commenters recommendation to delay reporting of
emissions. Please see the response EPA-HQ-OAR-2009-0923-0049-11. If EPA deems it
necessary to more accurately estimate emissions from previous years for the purpose of
informing future policy, then EPA will apply the updated emission factors to previous years
submissions.
1251

EPA is using emissions data from the Clearstone studies as the basis for the leaker factors
proposed in this rule. EPA requests comments on the use of emission factors from the Clearstone
studies.
Section VIII of this document details a comparison between the emission factors proposed for
Subpart W and the API Compendium. API is not opposed to using emission factors developed
from Canadian studies. However, emission factors specific to U.S. operations would be more
appropriate. API is open to working with EPA to develop leaker emission factors specific to US
operations. API also recommends that the reporting rule be flexible and allow the use of updated
emission factors as they become available. In addition, the emission factors provided for Subpart
W from the Clearstone studies are not in the cited reports. EPA should provide details on the
emission factors cited for an open and transparent review by industry.
Response: EPA disagrees with the comment on how to update emission factors. Please see the
response to EPA-HQ-OAR-2009-0923-1299-5. EPA disagrees that the emission factors in the
rule are not appropriate. EPA has documented the data in the Canadian studies and the
derivation of emission factors in the rulemaking docket (EPA-HQ-OAR-2009-0923) under
Revisions to Processing Leaker Emission Factors in Rule Table W-2 and the Greenhouse Gas
Emissions from the Petroleum and Natural Gas Industry: Technical Support Document (TSD)
found in docket (EPA-HQ-OAR-2009-0923-0027).

Emission Factor Development
DPL is concerned that the emission factors utilized to calculate fugitive emissions in proposed
Subpart W are dated and likely inaccurate. These emission factors are not only a decade old but
were based on equipment and testing facilities nearly two decades ago. The aged proposed
Subpart W emission factors do not account for more modern and tighter natural gas distribution
systems. Use of these emission factors for GHG reporting will only lead to inflated fugitive
emissions figures that are counterproductive to the vast resources utilized to achieve them.
Response: EPA disagrees that the emission factors in the rule are inaccurate. EPA used the best
publicly available data in todays final rule. Please see the response to EPA-HQ-OAR-2009
0923-1299-5.
Organization: Citizens Energy Group
Commenter: Ann W. Mclver
1252

Citizens also urges EPA to reconsider the use [of] emission factors that are out of date as a result
of technology innovations or other conditions. The use of unrepresentative factors may
overestimate GHG emissions associated with fugitives from equipment, placing an unrealistic
expectation on the total emissions reductions available from the sector. Citizens encourages EPA
to work collaboratively with organizations to update these emissions factors, and to provide
sufficient flexibility in the rules to allow the use of new or updated emission factors as they
become available.
Response: EPA disagrees that the emission factor used in todays final rule are unrepresentative
of current equipment and operations Regarding how emission factors may be updated in the
rule, please see the response to EPA-HQ-OAR-2009-0923-1299-5 for more information.

Therefore, we urge you to postpone requiring fugitive emission reporting for natural gas
distribution under Subpart W. Instead, AGA would like to finish the joint project we have
undertaken with EPA and others to conduct field testing and to develop updated, modern,
accurate emission factors for natural gas distribution. AGA and its members have already
conducted work in this area and developed some emissions factors that are not reflected in this
proposal. And we are committed to further improvements in emission factors.
Response: The emission factors that the commenter cited have not been submitted to EPA, and
this work is not scheduled to be completed in the near future and therefore EPA cannot wait to
finalize this rule after the factors are finished. Please see the response to EPA-HQ-OAR-2009
0923-1020-4 for further details. EPA used the best publicly available data to develop emission
factors for natural gas distribution facilities. Please see the response to EPA-HQ-OAR-2009
0923-0049-11. If new and improved emission factors become available, at an appropriate time,
EPA may re-propose applicable parts of subpart W. Please see response to comment EPA-HQ
OAR-2009-0923-1299-5.

Third, we are concerned that your plan to use emission factors that were developed over a decade
ago for another purpose by the Gas Research Institute, GRI, that were based on limited field
work conducted nearly 20 years ago beginning in 1992 were concerned about that because
many things have changed since then. Much of the old cast iron and steel pipe has been replaced
or lined with new plastic pipe.
1253
And modern plastic pipe has significantly lower fugitive emissions. Materials and construction
methods and practices for operations have changed as well and improved. As you know, over the
past decade, our members have been working with your EPA Natural Gas STAR program with
Roger Fernandez whos here some place in the audience. And were working through that
program to develop and implement practices that make our systems tighter. Well, that work has
been going on, but those emission factors have not changed in the meantime. Those old
emissions factors represent an old snapshot in time and they do not reflect reality now. And they
tend to vastly overstate emissions from modern, tighter natural gas distribution.
Response: EPA disagrees that any industry wide changes in the materials used to make or
replace pipes will produce a change in emission factors. The data used to develop these factors
track the pipe material; therefore, the emission factors are material specific.
EPA appreciates the participation of Partners in the Natural Gas STAR program. However, there
are significant differences between the Natural Gas STAR program and intent of todays final
rule. Partners in the program report reductions in emissions using non-standardized
methodologies resulting in data that cannot adequately inform future policy. EPAs intent with
todays final rule is to monitor and report greenhouse gas emissions from the petroleum and
natural gas industry using methodologies that balance burden with data quality, while providing
sufficient data to inform future policy.
EPA used the best available, public data to develop the emission factors in Table W-1 through
W-7 of todays final rule. Please see the response to EPA-HQ-OAR-2009-0923-1299-5 for
further details.

Commenter:
75 FR 18647-18650: Tables W-1 through W-8
Comment: WBIH strongly recommends that fugitive emission factors provided in the proposed
rule, Tables W-1 through W-8, be moved to a separate EPA guidance document, incorporated by
reference into the rule; and, recommends flexibility in the use of fugitive emission factors (leaker
factors, component counts, equipment specific factors, and field-specific factors (developed at
the field level) for emissions quantification at onshore petroleum and natural gas production, and
onshore natural gas transmission compression facilities and underground natural gas storage
facilities.
Emission factors for fugitive emissions are not consistent between current guidance documents,
creating significant potential for fugitive emissions data inaccuracies. However, the use of
emission factors has to remain flexible so factors can be applied to most closely match
component operation and are consistent with current air quality permitting programs.
1254
Additionally, emission factors should be provided in a separate guidance document which will
facilitate changes without the need for a rule change.
Response: EPA disagrees with the suggestion to remove the emission factors from subpart W
and placing them in another referenced technical document. If the emission factors are updated
in the referenced technical document, then Subpart W will still have to be re-proposed. As a
result, there is no efficiency gain to updating emission factors in a separate document versus
having the factors in the rule itself. EPA disagrees with the comment to allow reporters to
choose methods that are not specified in subpart W. Please see Section II.L of the preamble to
The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98) and the response to
EPA-HQ-OAR-2009-0923-1151-9.

Flexibility to use alternative best available emission estimation methods is needed to avoid
unnecessary burden while ensuring quality data. Noble recommends that the emission factors in
Subpart W tables should be moved to a separate EPA reference document that is incorporated
into the rule by reference.
Response: EPA disagrees with the recommendation to provide the emission factors in a
referenced technical document. Please see the response to EPA-HQ-OAR-2009-0923-1074-19.

In addition, Noble strongly recommends that the emission factors in Subpart W tables (e.g.,
Tables W-1 and W-2) should be moved to a separate EPA reference document (such as or similar
to AP-42 emission factors) that is incorporated into the rule by reference. Reopening a rule to
update emission factors would be difficult to accomplish and introduce unnecessary delays
resulting in outdated emission factors. A separate EPA reference document should be developed
and subjected to a peer review process. Subsequent updates will facilitate emission factor
improvements and refinements from new and improved data that are collected and compiled.
This reference document can be periodically revisited and updated to ensure that the best
available data and emission factors are being used.
Response: EPA disagrees with the comment on how to update emission factors. EPA used the
best publicly available data to develop emission factors for natural gas distribution facilities.
Please see the response to EPA-HQ-OAR-2009-0923-1074-19.
1255


Emission factor update - The emission factors (EFs) should not be included in the proposed rule
because they are mostly outdated and should be updated. Companies must be provided the
flexibility and given reasonable time to develop more accurate EFs using current technologies.
Response: EPA disagrees with the recommendation to not include emission factors in the rule
to allow for emission factor updates. Please see the response to EPA-HQ-OAR-2009-1299-5.
Commenter: Kim Dang
TABLE W-1 OF SUBPART WDEFAULT WHOLE GAS EMISSION FACTORS FOR
ONSHORE PETROLEUM AND NATURAL GAS PRODUCTION
1256
1257
TABLE W-2 OF SUBPART WDEFAULT TOTAL HYDROCARBON EMISSION

FACTORS FOR ONSHORE NATURAL GAS PROCESSING PLANTS
1258
TABLE W-3 OF SUBPART WDEFAULT METHANE EMISSION FACTORS FOR

FIELD GATHER AND/OR BOOSTING STATIONS AND ONSHORE NATURAL GAS
TRANSMISSION COMPRESSION
1259

UNDERGROUND Natural Gas STORAGE

LIQUEFIED NATURAL GAS (LNG) STORAGE
1260
TABLE W-6 OF SUBPART WDEFAULT METHANE EMISSION FACTORS FOR LNG

IMPORT AND EXPORT EQUIPMENT TERMINALS
1261

NATURAL GAS DISTRIBUTION
1262
TABLE W-8 OF SUBPART WDEFAULT NITROUS OXIDE EMISSION FACTORS

FOR GAS FLARING [see original in pdf attachment]
Response: EPA agrees, and in todays final rule, the title of the tables have been changed as
suggested by the commenter. EPA deems the changes necessary to clarify the industry segment
for which the emission factors are applicable.
Gathering pipelines are not required to report emissions in todays final rule; please see Section
II.F of the preamble to todays final rule for a response to this comment.
Reporting of coal bed methane well water production is not required in todays final rule. Upon
further analysis, EPA determined that this source is not significant enough to be included in
todays final rule. Please see the Greenhouse Gas Emissions from the Petroleum and Natural
Gas Industry: Background Technical Support Document (TSD) found in docket (EPA-HQ-OAR
2009-0923-0027). The emission factor has been removed accordingly.

The proposed calculation for gas gathering pipelines is impractical. EPA is proposing to
not include reporting of fugitive emissions from natural gas transmission pipelines due to the
dispersed nature of the fugitive emissions and the fact that once leaks are found, the
emissions are generally addressed quickly. EPA fails to use this same logic for natural gas
gathering pipelines, which are significantly more dispersed than transmission pipelines. Gas
gathering lines are generally much smaller diameter and typically operate at much lower
1263
pressures than transmissions lines, resulting in a lower potential for emissions. Some
gathering pipelines even operate on a vacuum. It is also important to note that many
gathering and processing companies have implemented robust programs to find and fix
pipelines leaks. Contrary to the transmission pipelines, EPA appears to propose that gas
processing plants and producers conduct a physical count of piping components on gathering
lines and use population factors to determine emissions. Operators then apply another factor
(scf/hour/mile) to calculate GHG emissions from the pipeline segments.
GPA estimates that there are over 250,000 miles of gathering pipelines in the gathering and
processing sector, and hundreds of thousands of meter and valve settings, that would require
physical component counts. Even more impractical, is the requirement to conduct
compositional analysis at these sites to determine methane and CO2 concentrations. Further,
operators would have to track blowdowns and changes in small meter runs and pipe segments
in these hundreds of thousands of insignificant locations to report GHGs as required by
proposed Subpart W.
Response: Gathering pipelines are not required to report emissions in todays final rule. Please
see Section II.F of the preamble to todays final rule for a response to this comment.
Lastly, for entities who must report emissions from gathering lines, the emissions should be
calculated simply as the product of miles of gathering lines times an appropriate per mile
emissions factor. This value should be reported as a total for each company, rather than
being reported on a source or facility basis. Lastly, to accommodate this reporting method,
the rule should allow companies to use an overall estimate of CO2 and CH4 content.
Response: Gathering pipelines are not required to report emission in todays final rule; please

Further, as a separate sub-sector, API recommends the use of the default emission factor for
gathering pipelines, as provided in Table W-1. Natural gas gathering pipelines are widely
dispersed, are generally much smaller diameter, and typically operate at low pressures, resulting
in a lower potential for emissions. Some gathering pipelines even operate on a vacuum.
Emissions from the gathering lines are a small fraction of the total 224 million metric tons
attributed to the OPGP sector. According to information from the Gas Processors Association
(GPA), the total length of gathering pipeline in the lower 48 (L48) states is about 250,000 miles.
1264
Using the calculation methodology in the proposed rule (emission factor of 2.81 scf of
CH4/hr/mile) the CO2eq emissions for all L48 gathering lines is about 2.0 million metric tons,
i.e., less than 1% of the EPA projected OPGP sectors GHG annual emissions. Therefore, a
default emission factor is appropriate.
Response: Gathering pipelines are not required to report emission in todays final rule; please
It is also noted that the emission factor for the gathering pipeline segments (2.81scf/hour/miles)
is derived from an unexplained total pipeline emission estimate of 6.6 Bscf. See TSD at 147-148.
GPA suggests that if EPA has already established this emission estimate, they should simply add
it to the rolled-up GHG inventory for the natural gas industry rather than requiring operators to
expend significant resources re-creating the exact same number.
Response: Gathering pipelines are not required to report emissions in todays final rule; please
Emission Factor Comparison to the API Compendium
API supports the use of emission factors from the API Compendium, as these are widely used
and accepted by the oil and natural gas industry worldwide. API compared the emission factors
provided in Tables W-1 through W-7 to the API Compendium. The following summarize this
comparison. Tables at the end of this section show the numeric comparison of EPAs proposed
emission factors versus API Compendium emission factors. Errors and recommendations for
specific emission sources are noted below.
General Comment
- The TSD shows a CH4 density being applied to emission factors that are on a TOC or THC
basis. This error should be corrected and would be avoided if the Subpart W emission factors
were provided on a mass basis.
Onshore Production
- For Onshore Production gas service, the TSD cites the GRI/EPA study (Methane
Emissions from the Natural Gas Industry. Volume 8. Tables 4-3, 4-6 and 4-24, June
1996) for the gas service fugitive emission factors for onshore production. These
1265
emission factors were compared to API Compendium Table 6-12, which cites EPAs
Protocol for Equipment Leak Emission Estimates (EPA-453/R-95-017, EPA Office of
Air Quality Planning and Standards, November 1995, Table 2-4). This comparison is
provided in Table VIII-1 below. The API Compendium emission factors are higher than
those provided in Table W-1 for valves and open-ended lines. The values are the same for
connectors. For these factors, the TSD combines the Western US and Eastern US data
from the GRI/EPA study to develop the emission factors. It would be more appropriate to
weight the geographical region factors by the representative portion of the total US
production so as not to bias any one geographical area. However, API supports using the
emission factors cited in the API Compendium, which are generally more conservative
than the proposed emission factors.
- Table W-1 labels the emission factors by service: gas, light crude and heavy crude. API
Compendium Table 6-12 presents fugitive emission factors by service type based on
EPAs Protocol for Equipment Leak Estimates (EPA-453/R-95-017). However, the
emission factors are more comparable to API Compendium Table 6-14 factors (from API
4615), which are based on facility type. The tables in Subpart W do not explain this, as
they refer to the factors as service based factors. This could lead to erroneous application
of the emission factors. Please refer to Table VIII-1 for the comparison of gas service
components, and Table VIII-2 for the comparison of light crude and heavy crude service.
Based on this comparison, it appears the emission factors in Table W-1 are
misrepresented as service-specific factors and should instead be described as facilityspecific factors. Here also, API supports using the emission factors cited in the API
Compendium.
- The light and heavy crude oil service emission factors from Table W-1 match the average
emission factors provided on pages 146 and 147 of the TSD (Appendix L), which cites
API, Emission Factors for Oil and Gas Production Operations (Table 9, page 10), API
Publication Number 4615, January 1995. However, when converting the total gas
emission factors to an scf basis, the TSD incorrectly multiplies the original API emission
factors by the CH4 content and then divides by the CH4 density. The original total gas
emission factors should instead just be divided by the natural gas density to calculate the
total gas emission factor from a mass basis to an scf basis. This error should be corrected
and would be avoided if the emission factors were provided on a mass basis. With this
correction, the emission factors should match those from the API Compendium.
Gas Processing
- The references cited in the TSD do not provide the emission factors shown in the TSD.
The TSD cites EPA. Identification and Evaluation of Opportunities to Reduce Methane
Losses at Four Gas Processing Plants. Clearstone Engineering Ltd. June 20, 2002.
www.epa.gov/gasstar/documents/four_plants.pdf and National Gas Machinery
Laboratory, Kansas State University; Clearstone Engineering, Ltd; Innovative
Environmental Solutions, Inc. Cost-Effective Directed Inspection and Maintenance
Control Opportunities at Five Gas Processing Plants and Upstream Gathering
Compressor Stations and Well Sites. For EPA Natural Gas STAR Program. March 2006.
EPA should provide details on the derivation of the emission factors provided in Table
W-2. In addition, API could work with EPA in the future to develop leaker emission
factors specific to US operations.
1266
The cited reports provide average emission factors, and do not break them out by
reciprocating compressor, centrifugal compressor, and other components. The
reports do provide detailed tables for all measurements conducted in the study, but not
with the information needed to derive the emission factors presented in the TSD. EPA
should provide details on the derivation of the emission factors provided in Table W-2.
Emission factors from API Compendium Table 6-15 (shown in Table VIII-2 below) cite
the same reference as the TSD. However, as noted above, a comparison can only be made
for the Other Components-Gas Service shown in Subpart W Table W-2. EPA should
provide details on the derivation of the emission factors provided in Table W-2.
Transmission, Storage, and Distribution

- The API Compendium does not provide leaker emission factors specific to transmission,
storage and distribution, so a direct comparison is not possible. API could work with EPA
in the future to develop leaker emission factors specific to US operations.
- For pneumatic controllers, Tables W-3 and W-4 presents the low-bleed emission factor
from onshore production. The API Compendium provides emission factors specific to
pneumatic devices observed in transmission and storage operations. The pneumatic
controller emission factors for transmission and storage should be corrected to apply
factors specific to those sectors rather than apply factors developed for production
operations. API recommends using the sector appropriate pneumatic controller emission
factors from the Compendium.
- The leaker emission factors provided in Tables W-4 and W-7 are taken from Appendix
K of the TSD (page 142) without references or derivations. Therefore, it is not possible to
evaluate these emission factors. EPA should provide details on the derivation of the
emission factors provided in Tables W-4 and W-7.
- The regulator and open-ended line emission factors provided in Table W-7 do not
match the emission factors shown on page 142 of the TSD. It appears that the regulator
emission factor from the TSD was assigned to the open-ended line in Table W-7. This
error should be corrected.
- As shown in Table VIII-3, the emission factors for the below grade M&R Stations,
distribution mains, and distribution services compare well with the API Compendium.
API supports the use of emission factors consistent with the API Compendium.
- The units shown for the distribution mains emission factors should be scf CH4/mile-yr,
not scf CH4/service-yr. This error should be corrected. With this correction, the emission
factors compare well with the API Compendium emission factors, as shown in Table
VIII-4.
LNG Operations
- The emission factors provided in Table W-5 are derived from oil and gas production light
liquid emission factors and are not specific to LNG operations. The API Compendium
does not provide LNG emission factors to compare with Subpart W. The API
Compendium notes (p. 2-16) that CH4 emission factors for LNG vents are not well
developed. LNG systems are designed to avoid contact with the outside air and great
effort is taken to prevent vented and fugitive losses. API recommends excluding emission
estimates for LNG operations until LNG-specific emission factors are developed.
- Table W-5 provides an emission factor for vapor recovery compressors in units of scf
CH4/component-hr. The methane emissions per compressor are from the EPA Inventory
1267
of U.S. Greenhouse Gas Emissions and Sinks: 1990-2007. The TSD shows the original
emission factors as 100 scfd CH4/compressor and applies a CH4 content of onshore light
crude associated gas that is actually the weight fraction of CH4 for light crude service
components in production. This error should be corrected.
Flare N2O Emission Factors
- Table W-8 provides N2O emission factors for flares, which match values provided in API
Compendium Table 4-11. However, the API Compendium notes that very little
information is available for N2O emissions from flares. The factors cited by Subpart W
and the API Compendium are from IPCC, which provides a 95% uncertainty bound of
10% to +1000% for these factors. Due to the inaccuracies and insignificant contribution,
API recommends that N2O emissions from flares be excluded from Subpart W.
- The units for sweet gas processing and sour gas processing should be tonnes/MMscf
of raw gas feed. This error should be corrected. However, due to the inaccuracies and
insignificant contribution, API recommends that N2O emissions from flares be excluded
from Subpart W.
Table VIII-1. Production Gas Service Emission Factor Comparison
1268
Table VIII-2. Production Crude Service and Processing Emission Factor Comparison
Table VIII-3. Distribution M&R Station Emission Factor Comparison
1269
Table VIII-4. Distribution Pipeline Emission Factor Comparison
Response: EPA agrees that methane density should not be used to convert total hydrocarbon
emission factors with units of pounds per component per day to standard cubic feet per
component per day. In todays final rule, the conversion was completed using an average
density for natural gas.
EPA disagrees with the comment to weight the onshore production emission factors. EPA used
the same data as API used to derive its emission factor for gas services, which is the GRI/EPA
study (Methane Emissions from the Natural Gas Industry, Volume 8, June 1996), However, EPA
determined that the emission factors and the operational characteristics for onshore production
facilities in the two geographical regions differ sufficiently to justify a separation. As a result,
the emission factors for onshore production in todays final rule are for the eastern and western
regions of the U.S. Please see the rulemaking docket (EPA-HQ-OAR-2009-0923) under
Equipment-Level Population Emission Factors for Onshore Production.
EPA clarifies that the use of service-type is synonymous with APIs use of facility type in
API Publication Number 4615. Thus, the text in Table W-1A remains unchanged in todays final
rule.
EPA agrees with the commenter that the API compendium does not have leaker emission factors
for the onshore natural gas transmission compression, underground natural gas storage, or natural
gas distribution segment. However, EPA used the best, publicly available data to derive its leaker
emission factors for these segments of the petroleum and natural gas industry. EPA provided
derivation of leaker emission factors found in Tables W-2, W-3, W-4, and W-7 in the rulemaking
docket (EPA-HQ-OAR-2009-0923). In the docket, EPA provided these data, a description of the
methodology employed, and the resulting emission factors. The leaker emission factors in Table
W-2, W-3, W-4, and W-7 in the final rule are not directly from the cited reports. EPA used the
raw data from these reports and a methodology different from these reports that tailors leaker
emission factors to the final rule. These methodologies, employed to derive the emission factors
in Table W-5, W-6, and W-7, are available in Greenhouse Gas Emissions from the Petroleum
and Natural Gas Industry: Technical Support Document (TSD) found in docket (EPA-HQ-OAR
1270
2009-0923-0027) and Revisions to Processing Leaker Emission Factors in Rule Table W-2 of
the rule making docket (EPA-HQ-OAR-2009-0923).
EPA agrees that the emission factors for regulators and open ended lines in distribution listed in
the Technical Support Document (TSD) found in docket (EPA-HQ-OAR-2009-0923-0027)do
not match those provided in Table W-7. EPA has revised the Background Technical Support
Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923) to match the
emission factors in Table W-7.
EPA agrees with the commenters suggested correction to Table W-7 in the proposed rule, and
todays final rule text was revised. The unit for distribution mains has been changed to standard
cubic feet per hour per mile of distribution main (scf/hour/mile).
EPA agrees that the API compendium includes a pneumatic device emission factor specific to
the transmission sector. However, this emission factor does not differentiate high-bleed
pneumatic devices from low-bleed pneumatic devices. EPA is requiring this level of specificity
from pneumatic devices to inform future policy. As a result, EPA is retaining the use of its
emission factors for pneumatic devices in the transmission sector in todays final rule.
EPA disagrees with the commenter on excluding LNG facilities and has retained the
requirements for reporting. LNG facilities in most cases handle natural gas and hence are liable
to equipment leaks just like in other gas handling facilities. EPA reiterates that the goal of the
Final Mandatory GHG Reporting Rule (Final MRR) (40 CFR part 98) is to quantify emissions
from major sources in the petroleum and natural gas industry. EPAs criteria are clearly outlined
in the Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry: Background
Technical Support Document (TSD) found in docket (EPA-HQ-OAR-2009-0923-0027) under
Selection of emissions sources for reporting. EPA used the best publicly available data to
develop emission factors in todays final rule. In addition, EPA will evaluate the LNG facility
data received through the MRR, which are based on emission factors and methodologies in
subpart W. EPA will also evaluate other emission factors that EPA may receive. EPA may also
consider using direct measurement of equipment leaks and vents if EPA deems the LNG facility
emissions data received through the MRR is insufficient to inform policy.
EPA agrees that the units for the emission factor for vapor recovery compressors should be
standard cubic feet per hour per compressor (scf/hour/compressor) and has revised it accordingly
in todays final rule. In addition, EPA agrees with the comment regarding the application of the
weight fraction of CH4 for light crude service components in production to convert the original
emission factor, and EPA has revised this by removing this weight fraction.
In todays final rule, EPA has replaced the proposed rules flare stack N2O emission factor with a
more accurate methodology to quantify N2O emissions; please see the response to EPA-HQ
OAR-2009-0923-1027-11. All emissions using this methodology will be calculated in units of
metric tons N2O emissions.
1271

Fugitive Emissions INGAA is Receptive to the Proposed Approaches but Revisions and
Refinements are Needed to Add Flexibility and Clarity, and Facilitate Improvements
For fugitive emission estimates from natural gas transmission and storage, the breadth of the
fugitive leak survey program, the basis for emission factors (EFs), and prescriptive technology
requirements are cause for concern. Although EPA has reduced direct measurement requirements
in the re-proposed Rule, it continues to mandate leak screening. INGAA is willing to accept the
survey approach and proposed EFs at this time, but several revisions and clarifications are
needed to make this a workable approach, including:
- Emission factors should be moved to a separate EPA technical document and
incorporated by reference to facilitate EF improvement;
- Alternatives to the optical gas imaging instrument (also referred to as optical camera)
should be allowed for leak screening;
- Practices and procedures for the optical camera developed for VOC leaks from other
industrial sectors should be revised to appropriately address methane leak surveys from
natural gas transmission and storage sources;
- Provisions for eliminating or decreasing the frequency of leak screening should be
adopted if data and/or EF advances meet accuracy needs or leak screening data indicates
that intervals can be increased; and,
- To eliminate ambiguity, definitions and clarification are needed for component types and
associated criteria for conducting a leak survey.
A. Emissions Factors Should Be Published in a Separate EPA Technical Document and
Incorporated By Reference to Facilitate Emission Factor Updates
INGAA strongly recommends that emission factors (EFs) in Subpart W tables (e.g., Table W-3
and W-4) should be moved to a separate EPA reference document that is incorporated into the
rule by reference. Reopening a rule to update emission factors would be difficult to accomplish
and is a time consuming activity, resulting in outdated EFs. Thus, a separate EPA reference
document should be developed and maintained. Review and updates would still require public
involvement and peer review, but this approach will facilitate EF improvements and updates as
new data are collected and compiled. This reference document can be periodically revisited and
updated to ensure that best available data and EFs are being used.
A supplemental technical document incorporated by reference provides a means to advance
state-of-the-science without requiring onerous rule revision / re-opening. Impeding more
accurate natural gas systems emissions reporting due to the rule revision process will preclude
timely updates and changes to the EFs necessary to advance the state-of-the-science and
facilitate informed decisions on fugitive emissions.
The Proposed Subpart W emission factors are questionable in some cases. For example the
compressor blowdown valve leaker EF is more than an order of magnitude greater than any
1272
component EF listed in Table W-3. This value is significantly higher than the value of 276.4
scf/hour per component listed in the 2007 Canadian Best Management Practices and 61.8
scf/hour per component listed in the INGAA Guidelines. A separate peer-reviewed EF document
that contains background data used to develop the EF should be provided for all listed
components.
Additional questions and concerns arise from the application of EFs that were developed using
methods (i.e., Method 21) to detect leaks other than the optical camera. No credible study or
technical analysis has been completed to support the application of the leaker EFs in Table W-3
and W-4 to leaks identified using the optical camera. The Method 21 approach and optical
camera have different sensitivities. These differences and the potential implications on fugitive
emissions estimate accuracy have not been reconciled. Therefore, it is conceivable that near-term
updates to EFs will be desirable based on either a more in-depth understanding of the EPAsanctioned EFs or an exponential growth of near-term data as rule requirements are
implemented.
EF studies have recently occurred and recently initiated studies are ongoing. INGAA, EPA, and
other stakeholders have been engaged in programs to improve EFs and such efforts should not be
undermined (e.g., University of Texas-Austin study, US EPA Cooperative Agreement no. XA
83376101UTx, which includes natural gas transmission participation). As discussed below, EF
projects should be undertaken with the goal and objective of improving current EFs and
ultimately simplifying or replacing annual leak surveys. For example, additional data could
document an alternative survey frequency or a need to focus on facility subsystems, such as
components in vibration or heat-cycle service that are more prone to leak.
In addition, Final Rule text and associated equations that reference EFs should be generically
written. For example, at a future point in time, alternative EFs may be desired that are
population-based to replace the currently-proposed leaker EFs (or vise versa). The separate EPA
technical document can provide the EFs and the context (e.g., activity data needs) for emission
calculations. Restricting some fugitive estimate approaches to leaker-only EFs and restricting all
emission factors to those published in the Rule will be too restrictive and will suppress
innovation and improvements in the accuracy of GHG emission reports.
Response: EPA disagrees with the commenters contention to move the emission factors to a
separate document. Please see the response to EPA-HQ-OAR-2009-0923-1074-19 for further
details. EPA disagrees with the recommendation to not include emission factors in the rule to
allow for emission factor updates. Please see the response to EPA-HQ-OAR-2009-1299-5. EPA
disagrees with eliminating or decreasing the minimum annual leak screening to calculate
emissions. Annual reports must be calculated and submitted each year; please see Section II.H
of the preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98).
EPA recognizes the prevalence of multiple methodologies for leak detection and has included
several additional leak detection techniques including flame ionization detectors, catalytic
oxidation/thermal conductivity detectors, and soap solutions as per Method 21; please see
Section II of the Preamble to todays final rule for further details.
1273
EPA disagrees that component definitions and leak detection methods are not in the rule. In
todays final rule, component definitions in leak surveys are in either Section 98.6 or Section
98.238 of todays final rule. Methods for using leak detection equipment are in Section 98.234
of todays final rule.
EPA supports studies conducted by all stakeholders to improve emission factors. If updated
emission factors become available through these studies, EPA may re-propose applicable parts of
subpart W. Please see the response to EPA-HQ-OAR-2009-0923-1299-5.
Upon further analysis and review, EPA determined there was insufficient data for Table W-3 for
transmission leaker emission factors for compressor blowdown valves. Todays final rule has
been revised and Table W-3 no longer has a compressor blowdown valve leaker emission factor.
In todays final rule, the valve leaker emission factor in Table W-3 is for block valves and
control valves. EPA has documented its derivation of emission factors used in todays final rule
in the Revisions to Processing Leaker Emission Factors in Rule Table W-2 memo.
EPA disagrees with the comment on the use of emission factors developed from studies using
Method 21 and the application of the optical gas imaging instrument in the rule. EPA chose the
optical gas imaging instrument in the Alternative Work Practice to 40 CFR parts 60, 63 and 65,
and in todays final rule, is also allowing use of Method 21 for components within reach of a
person standing on a fixed surface. The definition in todays final rule of a leak for Method 21
is 10,000 ppm by OVA/TVA, and the mass equivalent of that leak is 60 grams per hour. Since
the optical gas imaging instrument can easily detect leaks well below the AWP, it is qualified.
The explanation of the relationship between the 60 grams per hours leak threshold and various
leak definitions is provided in EPAs Preamble to the AWP, and in docket EPAHQ-OAR-2003
0199-0005.

Organization: Texas Commission on Environmental Quality
Commenter: Mark R. Vickery
The TCEQ concurs with the use of fugitive factors for estimating emissions from leaking
fugitives instead of direct monitoring to effectively reduce the emissions estimation and
reporting burden. Their use, rather than direct monitoring, is considered acceptable practice in
other industries. They may, however, be draft or out of date. A stated goal of the inventory is to
lay a foundation for future regulatory action. The data collected reflects significant effort on the
part of the reporting companies and may have significant impact on any future regulatory
activities. Thus, accuracy of data remains a high priority and the TCEQ urges the EPA to update
the factors in the rule as improved factors become available.
Response: EPA may re-propose applicable parts of the Subpart W, if EPA deems it necessary to
update emission factors. Please see the response to EPA-HQ-OAR-2009-0923-1299-5.
1274

APGA supports estimating emissions using population and leaker emission factors. APGA
supports EPAs proposal to allow fugitive methane emissions from LDC operations to be
estimated based on population and leaker emission factors rather than by direct measurement. As
EPA correctly notes, many of the leaks on natural gas distribution systems are on buried
pipelines where it would be difficult, if not impossible, to directly measure leakage rates. APGA
urges EPA to update these emission factors as more accurate data becomes available.
Response: EPA may re-propose applicable parts of the Subpart W, if EPA deems it necessary to
update emission factors. Please see the response to EPA-HQ-OAR-2009-0923-1299-5.

Organization:
Commenter: M. Harrison
Please allow the use of emission factors for compressor related components. EPA should not
require direct measurements for all gas compressors. As the project manager of the GRI/EPA
studies from the 1990's that are cited in the proposed rule, I am familiar with many of these
emission measurements. The direct compressor measurements EPA is requiring for five fugitive
and vented components at compressor stations are more technical, complex, and expensive to
measure than EPA has estimated. Some emission sources are not possible to individually
measure without physical modifications to the site. EPA should continue to fund research that
refines the key natural gas emission factors. EPA should not require the entire industry to make
measurements for a GHG reporting rule. EPA could allow for direct measurement as an
alternative to emission factors, if the reporter elects, in their sole discretion.
Please allow for the use of scaled emission factors, rather than more detailed modeling that the
proposed rule requires for sources like glycol dehydrators, amine units, and tanks. For the
reporting rule purposes, the accuracy of scaled emission factors should be acceptable.
Response: EPA disagrees that it has underestimated the complexity and difficulty of
quantifying emissions from compressor stations. EPA does not require complex physical
modification to compressor equipment and is allowing acoustic leak detection devices to
measure through valve leakages. Please see the response to comment EPA-HQ-OAR-2009
0923-1099-18. Compressors in the onshore petroleum and natural gas production segment of the
industry use an emission factor to calculate emissions. Therefore not all segments of the industry
are required to make measurements to determine emissions from reciprocating and centrifugal
compressors. For further information please see the rulemaking docket (EPA-HQ-OAR-2009
0923) under Compressor Modes and Threshold. Glycol dehydrators, onshore production
storage tanks and blowdown vents below set equipment thresholds can report emissions using a
population emission factor in todays final rule. However, those emission sources above the
1275
equipment threshold are required to estimate emissions using more rigorous methodologies. For
further information please see the response to EPA-HQ-OAR-2009-0923-1011-39, EPA-HQ
OAR-2009-0923-1060-10, and EPA-HQ-OAR-2009-0923-1061-12 for glycol dehydrator,
onshore production storage tanks and blowdown vents, respectively. With regards to comment
on funding research, EPA will consider participating in studies to improve emissions factors
based on the priorities and budget of the Agency. EPA has allowed for alternative methods for
data collection in certain cases where it was applicable and proven methodologies existed. EPA
will consider adding direct measurement options for other sources when additional data comes
available and at which time the Agency considers it appropriate in order to meet the goals of the
mandatory reporting rule. It is unclear to EPA what the commenter meant with the term scaled
emissions factors.
In 98.233(p)(3) CAPP recommends emissions be estimated using average emission factors and
annual hours for each of the operational modes. This is based on the large number of
compressors this section is expected to affect (about 33000), and that as many as 18 annual
measurements could be required for each (compressors typically have two to six compression
cylinders and a measurement is required for three operating modes for each cylinder), resulting
in a level of effort required to quantify these emissions that is out of balance with the overall
contribution to the inventory.
Response: EPA agrees that some emissions from compressor components in different operating
modes are similar and can be combined into a reporter based emission factor. Please see the
Providing simplified reporting methods based on engineering estimation or emission factors,
not direct measurement for small components or units, such as small compressors and
combustion devices, or low-throughput dehydrators and storage tanks;
Response: EPA agrees and has revised todays final rule to allow glycol dehydrators, onshore
production storage tanks, blowdown vents, and small external combustion units below a set
equipment thresholds to report emissions using a population emission factor in todays final rule.
Todays final rule has also been revised to allow reciprocating and centrifugal compressors in
onshore petroleum and natural gas production to use an emission factor. For further information
please see the rulemaking docket (EPA-HQ-OAR-2009-0923) under Equipment Threshold for
Tanks, Equipment Threshold for Dehydrators, Equipment Threshold for Blowdowns,
Equipment Threshold for Small Combustion Units, and Compressor Modes and Threshold.
1276

Organization: Indaco Air Quality Services, Inc.
Commenter: Touche Howard
Allow a facility emission factor for standard components at compressor stations.
Response: EPA disagrees that facility level average emission factors should be used for
compressor stations. Please refer to the response to EPA-HQ-OAR-2009-0923-1011-19.

Monitoring methods and emissions quantification for: Above ground meter regulators and gate
stations 98.232(i)(1)
Section 98.233 (q) requires a leak detection survey of connectors, block valves, control valves,
pressure relief valves, orifice meters, other meters, regulators, and open ended lines using an
optical gas imaging instrument at above ground meter regulators and gate stations. The leak
detection survey as proposed is time consuming and expensive. For a detail of issues on leak
surveys please see above comment #2 under General Comments on Subpart W. Our company
does not have an inventory of all eight components listed in 98.232(i) (1). We would have to
visit every meter and regulator to create a component list. To eliminate the burden of the leak
survey NMGC suggests applying facility-level emission factors for above ground meters similar
to what EPA is proposing for below ground meter regulators. This will eliminate the need to visit
hundreds or thousands of meter regulators and perform leak surveys with scarce and expensive
infrared cameras. In order to be able to track emission reductions there should be way to adjust
the emission factor of a facility due to good maintenance practices and/or replacement of known
high emitters with newer equipment.
Response: The commenter has misinterpreted the proposed rule. EPA disagrees that leak
detection is required at all meter and regulator stations. Please see the response to EPA-HQ
OAR-2009-0923-1065-4. EPA did not include facility level average emission factors because it
does not provide sufficient activity data to inform future policy. Please refer to EPA-HQ-OAR
2009-0923-1011-19.

1277

Reduce Number of Components Requiring Leak Surveys
As we contend elsewhere in these comments, EPA should either allow the use of facility level
emission factors or reduce the number of components requiring leak surveys. This will help
reduce compliance burdens and costs for underground natural gas storage facilities.
Response: EPA disagrees and did not include facility level average emission factors because it
does not provide sufficient activity data to inform future policy. Please refer to EPA-HQ-OAR
2009-0923-1011-19.
Section 98.233(q)(2) Leak detection and leaker emission factors for Onshore natural gas
processing facilities.
Currently, Section 98.233(q) requires surveying centrifugal compressor dry seals (for units
operated at processing facilities) with an IR camera. Given the fact that dry seals are designed to
emit in small amounts under normal operating conditions, it seems more reasonable to apply a
population count and population emission factor to this component rather than require leak
detection. Therefore, El Paso recommends that requirements for dry seal components fall under
Section 98.233(r).
Response: EPA does not agree that dry seal emissions should fall entirely under equipment
leaks in Section 98.233(r) of todays final rule. Dry seals may emit gas from two point sources:
at the seal face and at vents (where present). The seal face emission can be very small, depending
on the presence of an atmospheric pressure vent between tandem dry seals. EPA does not have
sufficient data to create an emission factor for dry seal emissions at the seal face, so while a
leak survey with infrared camera may detect emissions from the dry seal face, there is not a
factor to quantify such leakage in Section 98.233(r), and so, this emission will not be reported.
However, all vents must be surveyed, measured, and reported, including a dry seal vents. The
direct measurement of emissions vented from the dry seal contributes to the emission factors that
reporters must use to calculate emissions from various compressor modes. For more information
please see Compressor Modes and Threshold (EPA-HQ-OAR-2009-0023). EPA deems this
data quality is required to adequately inform future policy.
XVIII. Section 98.233(q)(3) Leak detection and leaker emission factors for Onshore natural gas
transmission compression facilities
1278
Table W-3 referenced under Section 98.233(q)(3) does not include a centrifugal compressor
component leaker factor for dry seals. Given this component was also excluded from the
transmission segment emission source list under Section 98.232(e) it is El Pasos interpretation
that the transmission segment does not have any monitoring requirements for centrifugal
compressor dry seals. El Paso requests confirmation of this interpretation.
If it was the EPAs original intent for this component to be included in the rule, El Paso
recommends the monitoring requirements be similar to those prescribed for processing facilities,
and that a population emission factor be added to Table W-3 for dry seals on centrifugal
compressor units operated at transmission compression facilities.
Response: EPA disagrees with the comments on transmission centrifugal compressors. In
todays final rule, measurement of vented emissions from a centrifugal compressor dry seal vent
pipe is required. EPA also disagrees with the use of a population emission factor to quantify
emissions from a dry seal compressor. Please see response to comment EPA-HQ-OAR-2009
0923-1011-46.
Table W-2: It appears that the emission rate per facility is the same for offshore and onshore
petroleum and gas production. Since different emission inventories were done for the onshore
and offshore emissions how can the average facility rate be the same? Please explain.
Response: EPA is unclear where the commenter observed that the emission rate per facility is
the same for offshore and onshore petroleum because Table W-2 in the preamble of the proposed
rule does not state as such. However, EPAs analysis concluded that the emission rate per
facility is not the same for offshore and onshore production facilities. Please refer to the
Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry: Background Technical
Support Document (TSD) found in docket (EPA-HQ-OAR-2009-0923-0027).
The Final Rule should clarify the count for mains in Table W-7.
As noted above EPAs method would presume that all mains, services and below ground M and
R stations are leaking, therefore the emission factors for these components would be multiplied
by the total number of services, the total number of below ground M&R stations and, APGA
assumes, the total miles of mains. Our uncertainty about the count for mains arises because of
footnote 2 to Table W-7 in the Proposed Rule which states that the count for mains would be
1279
units of scf/hour/service. Number of services does not apply when considering mains. APGA
believes that this is an error and that miles of main is what EPA intended to be used as the unit of
count. If this is correct, APGA asks that footnote 2 of Table W-7 be changed in the final rule to
reflect units of scf/hour/mile of main. If some other factor is to be used in the count for mains it
should be listed.
Response: EPA agrees with the commenters suggested correction to Table W-7 in the proposed
rule, and todays final rule text clarifies. The unit for distribution mains has been changed to
standard cubic feet per hour per mile of distribution main (scf/hour/mile).

a. Page 18650, Table W-7 of subpart W Default Methane Emission Factors for Distribution:
The footnotes and one heading of this table appear to contain errors and Laclede suggests they
should be revised as follows:
Population Emission Factor Distribution Services, Gas Service3
2 Emission Factor is in units of scf/hr/mile.
3 Emission Factor is in units of scf/hr/service.
Response: EPA agrees with the suggested unit change. Please see response to comment EPA
HQ-OAR-2009-0923-0955-3 for further details.
Development of Emission Factors
To develop emission factors, each key compressor component (blow down valve vent, rod
packing vent, and unit isolation valve pair) would be measured in the operating condition found
at the time of the measurement. Time dependant population emission factors would be calculated
for the mode in which each component was not surveyed, by averaging the leakage found in each
component category over the total number of components surveyed in that year. In addition to
the average leak rate (which is the population emission factor), the standard deviation, and the
minimum and maximum observed leak rates should be reported for each factor per component
category.
The following are the suggested categories for emission factors and are summarized in Table 4:
1280
1) Blow Down Valve Vent (combine Operating and Standby Pressurized modes)
2) Pressure Relief Valve Vent (combine Operating and Standby Pressurized modes)
3) Rod Packing Seal Vent(s) on Reciprocating Compressors
a. Operating Mode
b. Standby Pressurized Mode
4) Unit Isolation Valve Pairs Not Operating mode only
Table 4: Summary of Emission Factor Categories for Compressor Units
Although emission factors could be calculated by grouping data by units, facilities, operating
companies, or industry, El Paso recommends grouping data at the operating company level.
Grouping data by facility would most likely result in too few data points to be statistically valid,
while grouping data industry wide would increase the uncertainty due to wider differences in
equipment and operating practices.
Response: EPA agrees with the commenters suggested method to develop emission factors for
different operational modes. Please see response to comment EPA-HQ-OAR-2009-0923-1099
18.
Fugitive Emissions Sources
Finally, Subpart W requires all covered facilities to perform comprehensive emissions surveys of
their entire population of fugitive emissions on an annual basis. Should a leak be detected during
this survey, the current calculation methodologies specified in the proposal require operators to
assume that these fugitive emissions occur for the entire 365 days in the year. 75 Fed. Reg. at
18623. Those facilities required to estimate fugitive emissions based on population count (e.g.,
onshore production facilities) generally must make similar assumptions (i.e., that the emissions
occurred for the total time that the specific source associated with the fugitive emissions was
operating). See 40 C.F.R. 98.233(r), 75 Fed. Reg. at 18643. Because this is not the case with
regard to most fugitive emissions at most operations, these assumptions will result in a
significant overestimation of actual emissions that will artificially inflate the inventory. Indeed,
as USEPA acknowledges in the Preamble for Subpart W, "the petroleum and natural gas industry
is already implementing voluntary fugitive emissions and repair programs" for detected fugitive
1281
emissions that will result in correction of leaks. Id. at 18623. Nevertheless, IOGA-WV shares
USEPA's view that requiring more frequent emissions surveys would be both unduly
burdensome and impractical in light of the marginal levels of emissions that would be captured
in the inventory in light of the leak reductions and repairs that are undertaken pursuant to these
programs. If anything, IOGA-WV believes that the burdens associated with undertaking a
comprehensive annual emissions survey of these sources outweigh the benefits of including these
comparatively de minimis emissions in the inventory.
Response: EPA disagrees that equipment leaks are a small portion of emissions from the
petroleum and natural gas industry. Equipment leaks are a substantial percentage of emissions
from upstream production; EPA conducted detailed analysis in order to determine sources to
report in each segment of the industry. EPA does not agree that the burden associated with leak
detection is not justified by the benefits. Please see Section III.E of the preamble to todays final
rule for a description of the benefits of the rule. EPA recognizes the commenters concern that
assuming a leak duration of 365 days may overestimate emissions. Conversely, there will be
leaks that start after a leak survey is conducted and therefore an underestimation may occur as
well. Regardless, in todays final rule, EPA allows reporters the option to perform subsequent
facility-wide leak detection surveys and to adjust their emissions to account for components that
are subsequently found to be leaking or not leaking, respectively. EPA emphasizes that
adjustment of emissions is not allowed based on repair records alone. Reporters must assume
that a leaking component has been leaking starting from the beginning of the calendar year. In
addition, if only one leak detection survey is conducted during the calendar year, the reporter
must assume that the duration of the leak is 365 days unless the leak is fixed and a subsequent
official leak detection survey is conducted for an entire facility proving that the leak and others
were repaired. A goal with leak detection is to get a facility wide snapshot of equipment leak
emissions. If a reporter finds certain components leaking during a specific survey they are
usually not addressed immediately. During the time leaks are being addressed, other leaks will
appear which only subsequent facility wide leak detection surveys will ascertain.
General Comments Use of Direct Measurement
On page 18611 of the preamble, the EPA states, In this supplemental proposal, EPA is requiring
the use of direct measurement of emissions for only the most significant emissions sources where
other options are not available, and proposing the use of engineering estimates, emission
modeling software, and leak detection and publicly available emission factors for most other
vented and fugitive sources. We disagree that the EPA is requiring use of direct measurement of
emissions "for only the most significant emissions sources where other options are not
available." Direct measurement is required for transmission tanks, wet seal degassing,
reciprocating rod packing venting, flare stacks. Requiring direct measurement of transmission
tanks, wet seal degassing and reciprocating rod packing venting will require field personnel to
visit each site to determine which sources must be reported, and to measure each of those
1282
sources. Companies sending field personnel to even a portion of these sites will result in
emissions from vehicles, wear and tear on equipment, and valuable time lost by staff in the field
- none of which appears to have been taken into consideration when calculating the cost of
implementing this rule.
PAW requests that EPA reconsider the use of direct measurement for these sources after one
year. If these sources are less than 5% (in alignment with The Climate Registrys de minimus
emissions) of the total GHG emissions after the first reporting year, PAW recommends that EPA
propose emission factors and assumptions as an alternative to direct measurement.
Response: EPA disagrees that direct measurement of emissions is required at all transmission
tanks. EPA does not require reporters to schedule new trips to conduct direct measurement of
emissions. EPA disagrees with the commenter that emissions from transmission tanks, wet seal
degassing and reciprocating rod packing are de minimis in nature. Please see the response to
EPA-HQ-OAR-2009-0923-1060-12 for further details.

With respect to the following statement in 98.233(r)(2): In cases where the stream is almost all
CO2, the emissions factors in Table W-1 shall be assumed to be for CO2 rather than natural gas.
CAPP noticed that the emission factors in Table W-1 do not appear to be referenced to natural
gas and contend that it would be acceptable to simply use the factors (which were likely derived
from total hydrocarbon emission measurements) and speciate the resulting emission estimate
using the appropriate CO2 and methane content of the emitted stream. CAPP believes this
method would eliminate the uncertainty of the text in 98.233(r)(2).
Response: EPA agrees that the emission factors provided in Table W-1 of the proposed rule are
for whole gas. Todays final rule has been revised in 98.233 (r) (2) to require EOR operations to
use the reporters actual gas composition.

Table W-2
CAPP was unable to determine why two sets of emission factors are necessary to account for
equipment upstream of a de-methanizer and downstream of a de-methanizer. Fundamentally,
there is nothing different in the equipment and how it is operated other than the composition of
the gas and this is accounted for in speciating the total estimated emissions in equation W-18.
1283
CAPP is requesting clarification on whether or not the data support the division of emission
factors into these two categories, and if the presented emission factors actually taken from two
different populations.
Response: EPA agrees and in todays final rule the use of separate emission factors for
components before and after the de-methanizer has been eliminated. EPA has combined these
and developed a single emission factor. Please see the rulemaking docket (EPA-HQ-OAR-2009
0923) under Revisions to Processing Leaker Emission Factors in Rule Table W-2.

Provide streamlined emissions estimating methods for upstream production sites with
limited GHG emissions. For these sites, the API Compendium provides a reliable,
accurate, and streamlined method of estimating emissions. Upstream production sites
with CO2e emissions at or below 3,000 tpy based on the API Compendium would report
those emissions as calculated.
Require the largest-emitting upstream production sites to employ more rigorous
emissions estimating methods. Upstream production sites with CO2e emissions above
3,000 tpy based on the API Compendium would use the methods proposed by EPA (as
amended according to the additional comments provided below).
Response: EPA disagrees with the comment to allow reporters to choose methods from the API
Compendium to determine emissions from onshore production that are not specified in subpart
W. Please see Section II.L of the preamble to The Final Mandatory GHG Reporting Rule (Final
MRR), (40 CFR part 98). EPAs criteria for including emission sources under onshore
production segment for reporting are outlined in the Greenhouse Gas Emissions from the
Petroleum and Natural Gas Industry: Background Technical Support Document (TSD) found in
docket (EPA-HQ-OAR-2009-0923-0027). In addition, EPA set equipment thresholds for
onshore production tanks, dehydrators, and blowdown vents to focus the rigorous methodologies
on the largest contributors to that particular source. For further information please see the
rulemaking docket (EPA-HQ-OAR-2009-0923) under Equipment Threshold for Tanks,
Equipment Threshold for Dehydrators, Equipment Threshold for Blowdowns, Equipment
Threshold for Small Combustion Units and Compressor Modes and Threshold.

The agency can streamline the number of sources that must be monitored at individual facilities
1284
and emphasize engineering estimation approaches over direct measurement.

Response: EPA disagrees with this comment. Please see the response to EPA-HQ-OAR-2009
0923-1031-18.
The cost and burden of collecting direct measurement data seems to far outweigh any potential
gains in data accuracy. Overall, direct measurement methods should be removed and replaced by
engineering estimates using appropriate emissions factors, such as those found in the API
Compendium or similar sources.
0923-1031-18.

Quality of Measurement Guidance
API supports EPAs selection of a combination of direct measurement, engineering estimation,
and emission factors in the re-proposed Subpart W. We believe this mix of approaches will meet
EPAs stated intent to achieve the most cost effective coverage of emissions (EIA 1.3.2.1; pg.
1-11) and to collect a reasonable estimate of GHG emissions data that can be used to inform
future policy decisions (EIA 2.3 pg. 2-6) while addressing the problems with the methodologies
originally proposed in Subpart W. However, we have presented alternative methods to reduce the
cost and burden of the rule while still maintaining complete and accurate data. The methods are
detailed in a Section VII of this document.
In addition, API recommends moving the emission factors provided in Tables W-1 through W-8
to a separate document rather than including them in the regulation (API provides additional
comments on the specific emission factors in Section IV, response to item W15, and Section VIII
of this document). It may not be effective or efficient for EPA to provide emission factors in the
regulatory text. Citing the factors in the regulatory text restricts the flexibility to use new
emission factors as improvements become available and EPA resources will be required to
maintain and update the information. EPA and the oil and natural gas industry associations are
conducting measurement programs to improve emission factors. The rule may spur additional
activities. As a result, the rule should provide a mechanism for companies to use updated
emission factors as they become available.
1285
API requests:
- Flexibility should be afforded to reporters to use either direct measurements or
engineering analysis for all sources that require reporting under Subpart W.
- Flexibility should be afforded to reporters to use updated emission factors as they become
available. EPA should provide the emission factors in a separate document that is more
readily updated and maintained.
- EPA should further improve the methodologies and structure of the rule to meet EPAs
stated intent of balancing requirements and burden against emission data needs.
Response: EPA disagrees with the comment to allow reporters the option to choose to update
their emission factors. EPA disagrees with the comment on how to update emission factors.
Please see the response to EPA-HQ-OAR-2009-0923-1299-5. EPA disagrees with the
commenters suggestion to allow reporters the option for all sources to choose methods not
prescribed in the rule. Please see Section II.L of the preamble to The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98). EPA selected the monitoring methodologies
that would minimize burden on industry while maintaining the necessary quality and uniformity
of data to inform policy. Please see the response to EPA-HQ-OAR-2009-0923-1031-18.

Emission Quantification Methodologies
BP supports EPAs selection of a combination of direct measurement, engineering estimation,
and emission factors in the re-proposed Subpart W. We believe this mix of approaches will meet
EPAs stated intent to achieve the most cost effective coverage of emissions (EIA at 1.3.2.1;
pg. 1-11) and to collect a reasonable estimate of GHG emissions data that can be used to inform
future policy decisions (EIA at 2.3 pg. 2-6) while addressing the problems with the
methodologies originally proposed in Subpart W. However, given the significant burden which
we have estimated, we have included proposals to further reduce the cost and burden of the rule
while still maintaining complete and accurate data which are detailed later in this document.
BP recommends moving the emission factors provided in Tables W-1 through W-8 to a separate
document which the rule refers to rather than including them in the regulation itself. This would
enable more efficient updating of these factors as new emission factors improvements become
available. EPA and the oil and natural gas industry associations are conducting several studies
aimed at improving emission factors.
Requests:
- Flexibility should be afforded to reporters to use either direct measurements or
engineering analysis for all sources that require reporting under Subpart W.
1286
Flexibility should be afforded to reporters to use updated emission factors as they become
available. EPA should provide the emission factors in a separate document that is more
readily updated and maintained.
EPA should take all opportunities to further improve the methodologies and structure of
the rule to meet EPAs stated intent while balancing requirements and burden against
emission data needs.
0923-1151-9.
The rule should not require an actual fugitive component count. Due to the vast
expansion of the scope (i.e., inclusion of upstream facilities), this proposal has drastically
increased the burden of reporting and accounting for all the required source types. Anadarko
is particularly concerned with the requirement to count all the fugitive components on every
well site and small production facility that we operate. This requirement is very significant
in terms of effort and cost. Moreover, an actual count is not needed to develop a reasonable
estimate of fugitive emissions from leaking components. The total emissions from such
sources is a tiny fraction of total GHG emissions from this sector, so requiring an actual
count results in an estimate that is unnecessarily precise. Accordingly, we request that EPA
allow operators to generate an engineering estimate of the average component count for well
sites and other related upstream equipment.
Response: EPA agrees with the commenters concern that physically counting each component
in onshore natural gas and oil production operations would be overly burdensome. Please see the

The rule should not require an actual fugitive component count. Due to the vast expansion of the
scope (i.e., inclusion of upstream facilities), this proposal has drastically increased the burden of
reporting and accounting for all the required source types, AXPC is particularly concerned with
the requirement to count all the fugitive components on every well site and small production
facility that we operate. This requirement is very significant in terms of effort and cost.
Moreover, an actual count is not needed to develop a reasonable estimate of fugitive emissions
from leaking components. The total emissions from such sources is a tiny fraction of total GHG
emissions from this sector, so requiring an actual count results in an estimate that is
unnecessarily precise. Accordingly, we request that EPA allow operators to generate an
1287
engineering estimate of the average component count for well sites and other related upstream
equipment.
Response: EPA agrees with the commenter that physically counting components in onshore
natural gas and oil production operations would be unduly burdensome. Please see the response
to EPA-HQ-OAR-2009-0923-1151-126.

B. Recordkeeping and Reporting Requirements Should Be Limited to Data that Are Reasonably
Accessible and Pertinent to the Regulatory Objective
98.236 identifies reporting requirements associated with the sources listed in 98.233. Some
requirements are infeasible. In other cases, INGAA recommends that EPA not require data
collection based on the possibility that the data could be useful in the future, such as to formulate
new EFs or emission intensity factors.
INGAA comments follow for 98.236 where INGAA recommends deleting the item from the list
of reported parameters or including the parameter in the GHG Monitoring Plan required under
98.3(g)(5) rather than including the information in the annual report.
98.236(c)(9) Blowdown venting: Item (i) should not be required or could be reported in the
GHG Monitoring Plan. Item (ii) is not applicable as the compressor is driven equipment and not
the driver. Item (iii) should be clarified to indicate the volume associated with particular systems
that could be released through the vent, and this parameter is more appropriately documented in
the GHG Monitoring Plan.
98.236(c)(11) Tank emissions identified with optical gas imaging instrument: The first five
parameters i.e., (i) through (v) are input parameters associated with E&P Tanks software
emission calculations (i.e., see item (c)(10) list of parameters) and should be deleted from the
Rule. These parameters are not relevant when performing tank vent measurements from
malfunctioning scrubber dump valves.
98.236(c)(17) Centrifugal compressor wet seals: As noted in Comment IV.C, compressor
throughput is not readily available and reporting this parameter would add significant burden.
EPA has not identified why these data are needed. It is imperative that item (iii) is deleted.
98.236(c)(18) Reciprocating compressor rod packing venting: As noted in the Comment
V.D and the previous bullet, compressor throughput is not readily available and reporting this
parameter would add significant burden. It is imperative that item (i) is deleted. In addition, item
(ii) should be revised and reconciled with revisions recommended by INGAA above that address
multi-mode test requirements. Furthermore, items (iii) through (v) should be deleted from this
1288
section and included in the GHG Monitoring Plan rather than including the information in the
annual report.
98.236(c)(19) Fugitive emission sources using EFs: Item (i) should be clarified so that
appropriate counts are reported and unnecessary documentation is avoided. For example, as
written, item (i) could be interpreted to mean that all components must be counted and reported
even if the EF is leaker based. Item (i) could be clarified by adding (i)(A) and (i)(B) as
follows:
(A) Component count for each fugitive emissions source where component population is
reported for population emission factors, or
(B) Component count for each fugitive emissions source where the number of leaking
components are reported for leaker emission factors.
If EPA intended to require total component counts for leak surveys that use leaker EFs, INGAA
recommends deleting that requirement. Considerable burden is added to complete a total
component count, and transmission sector personnel with expertise to complete detailed facilitywide component counts are very limited.
98.236(d): The requirement for minimum, maximum and average throughput for each
operation in the natural gas transmission and storage segments is not clear. This requirement
should be deleted or these terms should be specifically defined. If this is intended to require gas
throughput values for compressor stations, INGAA recommends that the requirement be deleted
because such information is not readily available for many facilities. The implications from
deleting this requirement is not evident due to the lack of clarity in the requirement.
98.236(f): To avoid confusion and for consistency with requirements elsewhere in Subpart W,
this section should specifically identify the affected portable equipment based on the following
revision:
(f) Report emissions separately for production wellhead portable equipment.
Response: EPA reviewed the commenters concerns about the data reporting requirements in
98.236 for blowdown vent stacks and has revised the requirements in todays final rule. The
commenter has listed several reporting requirements for compressor blowdown vent stacks
which have been revised and are not included in todays final rule.
With regard to transmission tank emissions identified with optical gas imaging instrument, EPA
agrees that not all of the reporting parameters listed in the proposed rule were necessary to
determine emissions related to malfunctioning scrubber dump valves. Todays final rule no
longer requires the reporting of scrubber temperature and pressure, sales oil API gravity, tank
capacity, tank throughput, or tank control measures.
With regard to centrifugal compressor data reporting requirements, EPA disagrees that average
annual throughput should be eliminated. EPA requires the compressor throughput for analysis of
1289
the activity data and the resultant GHG emissions reports, as combustion CO2 will be
proportional to compressor throughput. EPA removed the minimum, maximum and average
throughput for each operation in todays final rule. Please see the response to EPA-HQ-OAR
2009-0923-1024-36 for further details.
With regard to multiple mode compressor testing, it was not EPAs intent to require that
reporters shutdown compressors annually for the purposes of monitoring. Todays final rule
clarifies that measurements are to be made in the as found mode, and that shutdown,
depressurized mode must be measured at least once every three years for each compressor. The
reporting requirements have been reconciled with this clarification. For more information,
please see the response to EPA-HQ-OAR-2009-0923-1099-18. Finally, EPA agrees that whether
or not each rod packing case is connected to an open ended line, the type of device used for
measurement, and the locations from where emissions were detected are not necessary data for
informing future policy. Todays final rule no longer requires that these data elements be
reported.
With regard to component counts corresponding to equipment leaks, EPA clarified the reporting
requirements after considering comments. In todays final rule, the number of components
found leaking during a leak detection survey is reported. However a complete component count
is required for components that use a population emission factor. Todays final rule has
eliminated the ambiguity of data reporting from these two sources.
With regard to data reporting requirements for portable combustion equipment, EPA agrees that
clarification was necessary such that portable combustion equipment reporting is for onshore
production equipment only. However, EPA disagrees that the data reporting requirements section
is the proper location for this clarification. EPA has clarified the combustion equipment that
needs to report under Section 98.232 of todays final rule, which lists all the emissions sources
that must report in each segment of the petroleum and natural gas industry. Todays final rule
defines that monitoring and reporting of combustion emissions from stationary and portable
equipment are required for reporting for onshore production and natural gas distribution.

98.233(q) Leak Detection and Leaker Emission Factors, and 98.233(r) Population Count and
Emission Factors:
a.) 98.233(q) allows the use of optical imaging to find leaks and then to apply emission
factors to estimate the emissions from the components found leaking. We request that this
method be extended as an alternative to onshore natural gas and petroleum facilities
which are proposed to be covered under 98.233(r). As API has ably pointed out,
1290
compliance with 98.233(r) could prove very cumbersome and burdensome due to the
difficulty and man-hours that will be required for accurate equipment counts
particularly for facilities as large as our North Slope operations. So we request the option
of using the method of 98.233(q) so we may avail ourselves of it if it proves more
efficient and feasible.
b.) We believe that the proposed rule contains an error. Both 98.233(q) & (r) have the
LNG Storage (98.232(g)(3)) and the LNG Import and Export Equipment (98.232(h)(4))
categories referenced. We don't believe that EPA intended to subject fugitive leak
components to both quantification methods so we request that EPA correct this.
c.) As an alternative to requiring the manual counting of facility components for fugitive
leaks, we request that EPA allow the use of facility component estimations as presented
in the API Compendium.
Response: EPA disagrees that it should allow leak detection in lieu of population count.
However, EPA has revised todays final rule for onshore production to allow equipment level
counts instead of component level counts. EPA deems the use of population emission factors to
be the least burdensome method for reporting emissions. Please see the Section II.E of the
preamble to todays final rule. In addition, EPA intends to maintain the use of population
emission factors so that consistent data is reported for onshore production.
EPA disagrees with the comment on duplication of emissions sources under Sections 98.233(q)
and 98.233(r) of todays final rule. Please see response to comment EPA-HQ-OAR-2009-0923
1061-5.
EPA disagrees that facility level average emission factors should be used. Please refer to the

API factors should be allowed for calculation of emissions from centrifugal compressor wet seal
degassing vents. Section 98.233 (o) would prescribe methods for calculating emissions from
centrifugal compressor wet seal degassing vents. EPA should modify the proposed rule so that
API factors may be used as an alternative if the operator so chooses. As indicated above, use of
API factors is standard in the industry and their use should be available as an alternative to
parties complying with Subpart W.
Response: EPA has reviewed this comment and disagrees. New data and increased knowledge
of industry operations and practices have highlighted the fact that centrifugal wet seal degassing
venting emissions estimates based on the factors used in API Compendium are outdated and
potentially understated. EPA determined that direct measurement of centrifugal wet seal
degassing venting is required to maintain the necessary quality of data to inform policy. Please
see the Technical Support Document (TSD) for todays final rule found in docket (EPA-HQ
OAR-2009-0923).
1291

Commenter:
EPA requests comment on the use of emission factors and ways in which these shortcomings
may be overcome. (page 69)
The EPA, under its Natural Gas STAR program, as well as regulated companies has conducted
direct measurements of methane leakage. With this data, most companies are aware of the
components that require direct measurement as well as those that can utilize default values. It is
appropriate to use emission factors for small components; however it is crucial to give
companies the ability to conduct engineering analysis so more relevant emission factors for
specific applications can be ascertained. It is important to give companies flexibility when using
emission factors so they can produce the most accurate result.
Response: EPA disagrees with the commenters suggestion to allow reporters the flexibility to
develop their own emission factor through engineering analysis. EPA selected the monitoring
methodologies that would minimize burden on industry while maintaining the necessary quality
and uniformity of data to inform policy. EPA may consider refinements in emissions factors
and, at an appropriate future date, update emissions factors as necessary. For additional
information on emissions factor updates, please see EPA-HQ-OAR-2009-0923-1299-5.

98.233(n)(7): Calculate N2O emissions using the emission factors for Gas Flares listed in Table
W-8 of this subpart.
OOC Comment: No equation is given to calculate N2O emissions from flares, and Table W-8
references an emissions factor in metric tons/MMSCF gas production or receipts. The emission
factor should apply to flared gas volume as CO2 and CH4 emissions do, rather than total gas
production or receipts.
Response: In todays final rule, EPA eliminated the use of flare emission factors based on
receipts. These emission factors are from the API compendium and are broad factors meant to
be applied to national level data. EPA deemed this level of reporting to be inaccurate for the
purposes of informing future policy. In todays final rule, flare N2O emissions are calculated
using an equation based on the volume of fuel that is combusted. This method, which can be
applied to specific combustion sources, significantly improves the quality of data gathered while
minimizing the burden on reporters.
1292

W11. (Preamble p. 69)
EPA considered several options for estimating emissions from fugitive emissions sources. One
option considered was to use a population count of each fugitive emissions source (e.g. source
types such as valves, connectors, etc.) and multiply it by a population emissions factor. EPA
requests comment on the use of emission factors and ways in which these shortcomings may be
overcome.
In order [to] minimize reporting burden, API members recommend that EPA allow the use of
engineering estimates to determine fugitive component counts. This would enable reporters to
apply average counts of fugitive components for typical equipment within a Sub-basin entity,
such as those provided in Section C.1.5 of the 2009 API Compendium (note the API
Compendium data are from Canada) or to develop typical component populations for their
operations, such as on a per site or per equipment basis. API presents an alternative method for
quantifying component counts in Section VII.N of this document.
Response: EPA disagrees with allowing engineering estimation as a methodology to determine
the population count of components. EPA revised todays final rule for equipment leaks in
onshore production to allow the option of major equipment level counts (wellheads, separators,
meters/piping, compressors, inline heaters, dehydrators) and apply default average component
counts per primary equipment. Please see Section II.E of the preamble to todays final rule.
However, in the other segments of the petroleum and natural gas industry, EPA requires
reporters to create an inventory of components by physically counting. This level of data quality
is necessary in these other segments of the industry to inform future policy, and incurs an
acceptable cost burden.

Fugitive Emissions Estimation
EPA considered several options for estimating emissions from fugitive emissions sources. One
option considered was to use a population count of each fugitive emissions source (e.g. source
types such as valves, connectors, etc.) and multiply it by a population emissions factor. EPA
requests comment on the use of emission factors and ways in which these shortcomings may be
overcome.
In order to quantify emissions without resulting in additional reporting burden, BP recommends
1293
that EPA allow the use of engineering estimates to determine fugitive component counts. This
would enable reporters to apply average counts of fugitive components for typical equipment,
such as those provided in Section C.1.5 of the 2009 API Compendium (note the Compendium
data are from Canada) or to develop typical component populations for their operations, such
as on a per site or per equipment basis.
Response: EPA disagrees with this comment. Please see the response to comment EPA-HQ
OAR-2009-0923-1151-72.

Paragraph 98.232 (i)(1) Natural Gas Distribution. Indicates that fugitive emissions from other
meters are to be reported. However, Table W-7 does not provide an emission factor for this
source.
Response: In todays final rule, EPA has removed other meters as an emission source in
natural gas distribution. An emission factor was not provided for this source in the proposed rule,
and was accordingly removed from 98.232 (i) (1) as suggested.

With this said, Laclede is once again compelled to state that applying standardized emission
factors to these system components, as with gas mains and service lines, will not yield accurate
fugitive emissions; it will only add unnecessary incremental cost to the utility and its ratepayers.
Moreover, system technicians visit these stations regularly to perform mandatory safety
inspections and prescribed maintenance, and the equipment at these stations is rarely found to be
leaking. In the rare instances when leaks are detected, they are promptly corrected in compliance
with the pipeline safety regulations.
Response: EPA disagrees that the emission factors in the rule are inaccurate. EPA determined
that distribution population emission factors would result in reduced burden on industry while
maintaining the necessary quality of data to inform policy. Please see the Technical Support
Document (TSD) for todays final rule found in docket (EPA-HQ-OAR-2009-0923). EPA
disagrees that the leak inspection requirements of the Department of Transportation meet the
requirements of subpart W. Please see the rulemaking docket (EPA-HQ-OAR-2009-0923)
Understanding the Substance of the DOT Regulations and Comparing Them to the Subpart W
Requirements.
1294

Limitations / misapplication of emission factors:
Fugitive emissions calculated using emission factors should not be the basis for requiring an
LDC distribution system to report under subpart W.
EPA's standard emission factors, multiplied by the number of system components covered by the
rule, will not result in a complete or accurate accounting of each LDCs gas losses. There are too
many other variables involved, such as gas line purges and blow-downs during construction and
repairs, other regular maintenance activities, and gas lost due to third-party dig-ins. Although
these are routine occurrences that LDCs are accustomed to managing and reporting them to
authorities, as appropriate, it is not practical to keep track of every such occurrence and attempt
to quantify the volume of gas associated with every release. For instance, not only would it be
cost prohibitive, but also highly unsafe to attempt to measure gas leakage from a broken line
while in the process of exposing it for the purpose of making repairs. In large, highly integrated,
looped distribution systems, it is not possible to accurately isolate the volume of such leakage
apart from other recognized categories of gas losses. Be assured, however, that gas releases of
every type are minimized due to the economic and safety aspects referenced above.
Also, the natural gas emission factors in use today, although dated, were developed to represent a
macro, nationwide, cross-section of industry applications. EPA actually stands to lose definition
by requiring operators to apply these factors on a micro, system by system basis because there
can be much variability in system construction and maintenance practices among system
operators, as well as marked differences in operating parameters.
Furthermore, because EPA is likely aware, the emission factors commonly in use today for pipes
do not differentiate between pipe diameters and the wide spectrum of internal gas line system
pressures, although they do take pipe material into account. EPA should recognize that
calculating fugitive emissions for a discrete distribution system by means of the published
factors would only by coincidence yield an accurate portrayal of that gas systems fugitive
emissions, and even then, there are no means by which to verify that the fugitive emissions so
calculated are correct.
Response: The commenter has misinterpreted the proposed rule. EPA did not include
distribution pipeline blowdowns or dig-ins, these sources did not fall under the 80:20 source
selection criteria. Please see the Greenhouse Gas Emissions from the Petroleum and Natural Gas
Industry: Background Technical Support Document (TSD) found in docket (EPA-HQ-OAR
2009-0923-0027). EPA disagrees that all distribution equipment leak emission factors are dated.
Please see response to comment EPA-HQ-OAR-2009-0923-1059-9. For further information on
the derivation of emission factors used in the distribution sector, please refer to the rulemaking
1295
docket (EPA-HQ-OAR-2009-0923) under Revisions to Processing Leaker Emission Factors in

Rule Table W-2 and Appendix L of the TSD (EPA-HQ-OAR-2009-0923-0027).
EPA disagrees with the comment on verification. Please see Section II.N of the preamble to The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98).
Representative sampling and measurements to develop emission factors / data to estimate entire
population emissions rather than test or sample every emission source. The large number of
onshore oil and gas production emission sources e.g. about 800,000 wellheads nationwide,
over 10,000 wellheads and 6,000 separators and storage tanks for Noble operations alone
preclude the cost-effective collection of required data and samples for each individual emission
source. For example, it is estimated that on the order of 20,000 man-hours would be required to
survey just the component counts in the Noble inventory, and components are a relatively small
emission source (estimated to be about 3% of the total Subpart W inventory in Table 2).
For each emission source, required parameters would be collected from a statistically random
sample of the emission sources in a basin. The average emissions determined for the emission
source (i.e. emission factor) would then be applied to each emission source in the basin to
calculate the total emissions estimate. As data objectives and policy are better defined, a more
robust data set will have been reported and serve as the basis for improving the sample size
determination and data needs. To provide clarity and compliance assurance, Noble recommends
that the maximum number of sources sampled each year in a basin would be 5% of the total or
30, whichever is less. The minimum number of sources sampled each year in a basin would be 5
or the entire population if less than 5.
Response: EPA disagrees with the comment regarding allowing onshore production reporters to
conduct their own random sampling to determine emissions, as this would not ensure the data
quality necessary to inform policy. Please see the response to EPA-HQ-OAR-2009-0923-1151
126. The commenter provided no details on how a random sample would be chosen, or what
parameters would be collected, or how emissions would be calculated. In addition the
commenter suggests that EPA needs to better define its objectives before it finalizes data
requirements for subpart W. EPA disagrees with this comment, as EPAs authority to
promulgate this final rule is well defined. Please see Sections II.C and II.Q of the preamble to
The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98)
Organization: Citizens Energy Group
Commenter: Ann W. Mclver
1296
Additional clarification is needed for the types of components listed in the other category
referenced in Table W-5 and Table W-6.
Response: EPA disagrees, as the proposed rule has defined the other category in Table W-5 and
Table W-6 as any component other than connectors, pumps, or valves, with footnotes within the
tables themselves.

Dow Comments that EPA's Proposal for Use of Leaker Emission Factors for Pressure Relief
Valves and Compressor Blowdown Valves in Table W-4 Has the Potential to Create
Uncertainties for the Regulated Community.
Dow operates an underground storage facility where the vent piping associated with a number of
Pressure Relief Devices and Compressor Blowdown valves is combined into a larger piping
network with a single vent opening to the atmosphere. If a leak is detected at the outlet of the
common discharge piping, it may not be possible to determine which pressure relief valve or
blowdown valve is leaking or even if more than one source is leaking in these cases. Thus, Dow
comments that EPA's final rule should contain one or both of the following options for this
situation where the vent piping is combined together:
- Allow the use of engineering analyses or additional measurements, if possible, to
determine which pressure relief valve, blowdown valve, and/or how many pressure relief
valves or blowdown valves are leaking or;
- Rely solely on the use of Population Emission Factors for calculating fugitive emissions
from pressure relief valves and blowdown valves and not require an annual survey for
these sources.
Response: EPA disagrees with the use of an engineering analysis to quantify venting emissions
from through-valve leakage. These methods will not accurately quantify venting emissions from
a significant source of emissions to adequately inform future policy decisions. However, todays
final rule has been revised and EPA is allowing the use of an acoustic leak detection instrument
to detect and measure venting emissions from through valve leakage from transmission tanks,
and centrifugal and reciprocating compressor venting. EPA disagrees with solely using
population emission factors for underground storage facilities. Please see Section II.E of the
preamble to the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002).

1297

Dow Comments that Regulated Entities Should be Able to Take Credit for Repairing Leaking
Equipment.
EPA solicits comment on whether an alternative approach that requires multiple surveys during a
year should be required in order to improve the quality of the emission estimate for leaking
equipment. Dow comments that EPA's final rule should continue to require a survey on an
annual basis, but that regulated entities should be able to take credit for repairing leaking
equipment during the year. For example, if a component is found to be leaking methane on July
15th and the regulated entity makes repairs to the component and stops the leak on July 16th,
then credit should be taken for the repair. In this example, the emissions would be calculated
using the "Leaker" emission factor from the period January 1 - July 15th, and using the
Population Emission Factor from July 16th to December 31st. It is our belief that almost all
regulated entities will address discovered leaks, thus this type of approach will increase the
quality of the emission estimates.
Response: EPA agrees to allow multiple leak detection surveys. Please see the response to
EPA-HQ-OAR-2009-0923-1014-9.

The first year level of effort to meet the required leak detection and monitoring of the
distribution system will be the most burdensome as technicians require training on leak detection
methods and the database of sources is established. Under the proposed rule, an annual database
of leaking components at above grade M&R stations would need to be developed to calculate
GHG emissions utilizing the leaker emission factors in Table W-7. We would likely build on
existing databases and drawings to help develop a compliance plan for monitoring leaks and
reporting emissions from above grade M&R stations. However, the additional information needs
would not likely be met by existing inventories or databases, and significant additional work may
be required depending on the number of leaking components. For example, MichCon already
maintains a database of components that lists the regulators, meters, pressure relief valves and
line heaters at each M&R station as required by the Michigan Public Service Commission. This
database contains information on the component manufacturer, model number, and design
operating pressures. There are engineering schematics for each station, but these designs will not
necessarily include all connectors, and some bypass lines will not be included in the schematics.
Because of the scale of distribution systems and the implementation issues described above, DTE
Energy supports delaying the implementation of greenhouse gas reporting for M&R stations for
at least a year. In place of leak detection, DTE Energy supports using default emission factors for
above grade M&R stations as is allowed for below grade M&R stations.
1298
Response: EPA disagrees with delaying the implementation of todays final rule. Leak detection
is not required at all meter and regulator stations, thereby reducing the burden on reporters.
Please see the response to EPA-HQ-OAR-2009-0923-1065-4. A leak detection survey can be
conducted anytime during the year, which provides additional time for training personnel and
developing databases. EPA does not anticipate the need for more time but with sufficient
documentation, the reporter may apply for best available monitoring methods so long as the
reporter meets very specific circumstances, as outlined in the best available monitoring methods
criteria. For further information please see Section II.F of the preamble to todays final rule.

Direct Measurement vs. Emission Factors
The EPAs use of emission factors derived from the GRI (Gas Research Institute)/EPA study
performed in 1996 do not take into account the variations occurring from new maintenance
practices, better sealing materials and an overall bigger concern for the fugitive emissions
resulting in lost and unaccounted for gas.
To simply categorize each of the leaking components by emission factor limits the ability to truly
quantify those emissions and create an accurate, up to date emissions inventory. Through
extensive field work and thousands of accurate leak measurements, it has been proven that the
use of the Hi Flow Sampler for direct measurement in all phases of compliance for the EPAs
Mandatory Reporting Rule 40 CFR 98 Subpart W will be the most accurate method available
today. It is estimated that the use of the Hi Flow Sampler for quantification use will cover
approximately 98% of all vented and fugitive emissions encountered during the emissions
inventory survey, making this a very effective and accurate device for conducting the surveys.
Whats more, the Hi Flow Sampler takes into account temperature and pressure at the point of
measurement which lends significant credibility to this measurement data thereby reducing error
associated with other estimation techniques. The balance of the quantification events may be
accomplished using a variety of other means including but not limited to:
1. Calibrated vent bag system
2. Rotameter
3. Anemometer
4. Ultrasonic flow device
The use of the Hi Flow Sampler for the direct measurement (quantification) of vented and
fugitive methane emissions has been approved as standardized methodology by both the EPA
and the United Nations FCCC (Framework Convention on Climate Change AM0023) for the
Kyoto Protocols Joint Implementation and Clean Development Mechanism programs. This is
proven technology that has demonstrated remarkable accuracy as verified in side by side
comparative studies with both rotameters and calibrated vent bags.
1299
Based on these facts, it is recommended that the Hi Flow Sampler, as a proven quantification
device be utilized to provide the most accurate, up to date data in emissions inventorying. This
will provide the best quantified emission inventories for compliance to the new EPAs
Mandatory Reporting Rule.
Response: EPA disagrees that it should require reporters to submit measured emissions data
with the high volume sampler for all equipment leaks and vented emissions. EPA is allowing the
use of the high volume sampler for venting emissions from centrifugal and reciprocating
compressors, where direct measurement to quantify emissions is required to adequately inform
future policy.

Fourth, we are worried that you are trying to go too far down into the weeds at distribution
facilities asking for estimated emissions from too long a list of individual components. Our
members do not typically keep a comprehensive inventory of the number of components that
they have at any given metering and regulator station. For example, one of our companies Ill
get to that in a moment. The proposed rule apparently would require knowing the number of
each type of component in order to apply a component-specific emission factor multiplied by the
number of components. And if we dont already have that list then were going to have to go out
to each of those stations out at remote locations and make an inventory.
Now, I know that this sounded logical in theory, but in practice, it would be onerous. It would
require our gas utility personnel to visit thousands of remote facilities surveying equipment for
each station in minute detail just to develop these lists of components. One company for example
operates at least 10,000 M&R stations. This will require many hours of work at great expense to
go out and do these surveys.
And then for what result? Because in the end all we can do with the old emission factors is to
calculate a fictional number that we all know is seriously inaccurate and is likely to vastly
overstate and misrepresent the level of actual emissions at distributional facilities.
Response: EPA disagrees with the comment that todays final rule requires creation of an
inventory of all components at meter and regulator stations. EPA is requiring leak detection and
a count of leaking components only at custody-transfer gate stations, which significantly reduces
the burden. Please see the response to EPA-HQ-OAR-2009-0923-1065-4 for further details.
EPA disagrees that emissions factors used in todays final rule are inaccurate and
misrepresentative. EPA used the best available, public data to develop the emission factors in
the natural gas distribution segment. Please see the response to EPA-HQ-OAR-2009-0923-1299
1300

The proposed rule does not provide any simplified method for determining applicability. Similar
to the final version of Subpart C - facilities with less than 30 mmBtu/hr aggregate maximum heat
rating are not required to report if only potentially subject to Subpart C - there should be a lesscostly method for determining applicability than requiring all of the direct monitoring,
engineering estimation, component counts, etc. that are required by Subpart W. Please consider
providing emission factors that would apply only to applicability determination calculations or
other less costly applicability determination method.
Response: EPA intends develop a screening tool to assist in the determination of which entities
are required to report under subpart W of 40 CFR part 98. Please see Section II.F of the preamble
to todays final rule for a response to this comment.

Section 98.233(r) Population count and emission factors. Table W-3 provides the emission factor
for onshore natural gas transmission natural gas pneumatic low bleed device venting. Paragraphs
Section 98.233(r)(2) through (7) reference Tables W-1 through W-7 for source category
emission factors. However, Section 98.233(r)(2) through (7) does not reference Table W-3 for
onshore natural gas transmission emission factors. In addition, the sources Section 98.233(f)(5),
(g)(3), h(4) are included in both Section 98.233(q) and Section 98.233(r), which is duplicative
and the sections have different requirements.
Response: EPA agrees, and has revised todays final rule to remove the reference to pneumatic
devices from Section 98.233(r) of todays final rule. EPA disagrees with the comment on
duplication of emissions sources under Sections 98.233(q) and 98.233(r) of todays final rule.
For further details, please also see the response to EPA-HQ-OAR-2009-0923-1061-5.

Suggestion 5: Allow leak measurements to be made as an option to determine bleed rates of
pneumatic control devices.
1301
Sections 98.233 (a) and (b) require the use of either manufacturers data for high bleed
pneumatics or the application of an emission factor for low bleed pneumatics. However,
identifying the model of a particular device may be difficult if name plates are obscured by paint
or wear. Additionally, some units may have been modified over time to reduce their bleed rate,
but accurate records of these modifications may not be available. As an option, the Hi-Flow
sampler or similar device could be used to make a measurement of the bleed rate from pneumatic
devices. If companies already have this technology in house or are having it contracted for other
measurements at their sites, this could provide a method to obtain data that might otherwise be
hard to get.
Response: EPA disagrees that it should allow reporters to submit measured emissions data for
pneumatic devices in todays final rule. EPA has revised todays final rule to require population
emission factors for pneumatic devices and pumps, to reduce burden. Please see Section II.F of
the preamble to todays final rule. In the proposed rule, EPA required reporters to determine the
manufacturer model of each device in their facilities, and then estimate emissions based on
manufacturer data. However, many potential reporters commented that the cost burden was too
high. EPA accepted these comments and revised the rule to population emission factors to
reduce compliance burden on industry while maintaining the necessary quality of data to inform
policy

Tracking natural gas bleed rate during normal operation for every high bleed device will be
difficult to compile for the thousands of devices. PAW asks the EPA to propose a reasonable
high bleed emission factor to use in place of manufacturer's data if manufacturers data is
unavailable. Further, it is unclear what the EPA means by "continuous." If it is a "continuous"
bleed device, why is there a time element in the equation? Lastly, it is unclear at what point is the
unit considered not in operation and therefore exempt.
Response: EPA agrees, and has revised todays final rule to include population emission factors
for pneumatic device venting, which will eliminate the need to determine the device
manufacturer model; please see the response to EPA-HQ-OAR-2009-0923-1011-8. In todays
final rule, all natural gas pneumatic devices are assumed to be venting during the entire reporting
period; please see the response to EPA-HQ-OAR-2009-0923-1039-8. The definitions of
continuous and intermittent bleed pneumatic devices are in Section 98.6 of todays final rule.
Natural gas driven pneumatic pumps: Noble recommends the use of an emission factor(s) (such
1302
as the emission factor(s) for natural gas driven pneumatic pumps listed in the API Compendium)
based on natural gas driven pneumatic pump counts in each reporting area to determine
emissions from pneumatic pumps. If EPA does not support the use of the emission factor
method, then Noble supports APIs suggested alternative methods for natural gas pneumatic
pumps.
Response: EPA agrees, and has revised todays final rule to use a GRI emission factor for
pneumatic pumps. Please see the response to comment EPA-HQ-OAR-2009-0923-0055 -11.

Natural gas pneumatic low bleed device venting: The sources covered by this section are also
covered by Section 98.233(r) for using population count and emission factors. We request that
EPA delete this section to avoid double reporting or, alternatively, allow an operator to use an
engineering estimate of the count of low bleed pneumatic devices.
Response: EPA agrees and has revised todays final rule. Please see response to comment EPA
HQ-OAR-2009-0923-1151-44.
Natural gas pneumatic high bleed device venting: We request that EPA allow an operator to use
an engineering estimate of the count of high bleed pneumatic devices and a typical bleed rate to
calculate emissions from this source category on a facility-wide basis rather than per each
device.
Response: In regards to engineering estimates of pneumatic device counts, please see response
to EPA-HQ-OAR-2009-0923-0582-23. EPA has included population emission factors to reduce
the burden of determining the bleed rate of each high-bleed pneumatic device required to report.
For further details, please see Section II.F of the preamble to todays final rule.
Table W-2 is missing a Low Bleed Pneumatic Device emission factor, although this table is
referenced for the emission factor in 98.233(b), Equation W-2.
1303
Response: EPA agrees, and in todays final rule, EPA has revised the rule and included a
population emission factor for low-bleed pneumatic devices that is applicable to the onshore
natural gas processing sector.
Sampling and Audit Program
The dramatic underestimates of emissions from the oil and gas sector, acknowledged by EPA,
demonstrate the critical need to immediately improve the quality of emissions factors used where
direct measurement is not required and the importance of providing well-audited emissions data
to support quality policy-making. To that end, EPA should commit to a rigorous periodic
statistical sampling and audit program, beginning immediately, in order to ensure continuous
improvement in the data collected under the MRR. EPA should establish, either as part of this
rulemaking or in a separate sample and audit program, a specific plan and detailed timeline for
carefully measuring emissions and maximizing the precision and accuracy of its estimation
protocols.
Such an audit program should not, in any way, delay establishing reporting requirements for this
sector and indeed should build on the data collected under the initial version of the rule. Rather,
the goal of such a program should be stepwise improvement using the data gathered from the
reporting rule in 2011 as the input for a sampling and audit program aimed at improving the
quality of the data reported in subsequent years, perhaps as early as 2012.
Because uncertainties in emissions estimates could result in large-scale data quality issues if they
become ingrained, EPA must commit to actively conduct sampling and audit to minimize the
known uncertainties in quantifying emissions from this complex and fragmented sector. Because
of the fragmented nature of this sector, there will be many factors to consider in constructing a
useful audit program. EPA should place a high priority on those sources with the least certain
reporting methods (e.g., pneumatic devices and onshore production and processing storage tanks,
metering of vent flows from tank emissions) and on those sources where wide variability in
emissions has been demonstrated, such as well completions. EPA should consider the use of
randomized samples, collected at varying times and locations, from sources with the highest
potential emissions and should place an emphasis on the top emitting source types. Guidance on
sources to target may be available from Californias current efforts to improve their O&G
inventory, which is planned for completion this summer. 185 In addition, EPA should work
actively with industry and equipment vendors to collect the broadest pool of direct measurement
data possible. Another available resource for helping determine the highest priority sources
would be the Western Regional Air Partnerships recent screening inventory analysis for the oil
185
The California Energy Commission has a contract to develop new natural gas system emission factors and the
California Air Resources Board has a contract to develop new emission factors for oil fields. The timeline for both
inventories is for completion in summer 2010. See Oil and Natural Gas Production, Processing and Storage Public
Workshop, CARB, Sacramento, California, December 8, 2009.
1304
and gas exploration and production sector, which was designed to help rank GHG source
categories associated with exploration and production operations. 186
It is possible that some sources are already collecting samples that could be used as part of
EPAs auditing program (e.g., in order to meet reporting requirements for ozone nonattainment
area plans). In these cases, EPA could simply designate a certain percentage of samples for
independent analysis and review. For those industry sources where direct measurement will not
be required, alternative requirements for independent auditing of emissions estimate methods
should be included as part of the final rule for those sources.
Response: EPA has reviewed these comments and disagrees that this rule should be fashioned
specifically to serve several additional purposes beyond its primary intent of informing future
policy. EPA had to balance cost burden on the industry with the quality and quantity of data
necessary to address future policy, without biasing the data toward any particular policy. While
direct measurement data is always very useful, this adds substantial cost. Much of the most
useful and necessary data collected in todays final rule is the activity data: numbers of sources
or frequency and duration of gas venting. Where publicly available, relatively recently collected
data can be used for emission factors; this is a cost savings to industry. In cases where no
credible, broad emissions data is publicly available, and emissions vary widely from site to site
or by operating practice, EPA had to include the cost of direct measurement. The recent studies
cited by this commenter, the Western Regional Air Partnership, and California Energy
Commission, were not publicly available in time for vetting in the proposed rule. EPA may
consider pertinent new data in the future as it becomes available. EPA cannot comment under
this rule on any other programs to establish a separate sample and audit program aimed at
maximizing the precision and accuracy of emissions estimating protocols. This is a policy
outcome that may be considered based on the results of the MRR.
A Robust Sampling and Audit Program is Needed to Ensure Continuous Improvement in Oil and
Gas Reporting.
The dramatic underestimates in emissions from the oil and gas sector demonstrate a need to
improve the quality of emission factors used in cases where direct measurement is not required,
and the importance of providing well-audited emissions data to support quality policy-making.
To that end, EPA should commit to a rigorous periodic statistical sampling and audit program,
beginning immediately, in order to ensure continuous improvement in the data collected under
the MRR. EPA should establish, either as part of this rulemaking or in a separate sample and
audit program, a specific plan and detailed timeline for carefully measuring emissions and
186
39 Oil and Gas Exploration and Production Greenhouse Gas Protocol Task 2 Report on Significant Source
Categories and Technical Review of Estimation Methodologies - Final Report, April 22, 2010. Available online at
http://www.wrapair.org/ClimateChange/GHGProtocol/docs.html.
1305
maximizing the precision and accuracy of its estimation protocols.

Such an audit program should in no way delay the issuance of reporting requirements for this
sector. Rather, the goal of such a program should be to concurrently design and implement the
program based on current estimates and then use the 2011 submittals to further refine the
process, perhaps as early as 2012.
Because uncertainties in emissions estimates could result in large-scale data quality issues if they
become ingrained, EPA should minimize uncertainties through sampling and auditing. The
fragmented nature of this sector suggests that there will be many factors to consider in
constructing a useful audit program. EPA should place a high priority on sources with the least
certain reporting methods (e.g., pneumatic devices and onshore production and processing
storage tanks) and on sources where wide variability in emissions has been demonstrated, such
as well completions. EPA should consider the use of randomized samples, collected at varying
times and locations, from sources with the highest potential emissions, and should place an
emphasis on the top emitting source types.
Guidance on sources to target may be available from Californias current efforts to improve its
oil and gas inventory, which are planned for completion this summer. 187 Another available
resource for helping determine the highest priority sources could be the Western Regional Air
Partnerships recent screening inventory analysis for the oil and gas exploration and production
sector, which was designed to help rank GHG source categories associated with exploration and
production operations. 188
Some sources may already be collecting samples that could be used as part of EPAs auditing
program (e.g., in order to meet reporting requirements for ozone nonattainment area plans). For
these sources, EPA could simply designate a certain percentage of samples for independent
analysis and review.
Response: EPA disagrees that this rule should be fashioned specifically to serve several
additional purposes beyond its primary intent of informing future policy. Also, EPA cannot
comment under this rule on any other programs to establish a separate sample and audit program.
For more information, see the response to EPA-HQ-OAR-2009-0923-1155-12.
187
The California Energy Commission has a contract to develop new natural gas system emission factors and the
California Air Resources Board has a contract to develop new emission factors for oil fields. The timeline for both
inventories is for completion in Summer 2010. See Oil and Natural Gas Production, Processing and Storage
Public Workshop, CARB, Sacramento, California, December 8, 2009.
188
Oil and Gas Exploration and Production Greenhouse Gas Protocol Task 2 Report on Significant Source
Categories and Technical Review of Estimation Methodologies - Final Report, April 22, 2010. Available online at
http://www.wrapair.org/ClimateChange/GHGProtocol/docs.html.
1306

Emission Factors
Considerable expense and time would be required by operators for monitoring, lab analyses, and
component counting to implement the proposed rule. Much of the information collected will not
vary substantially from year to year and specific emission factors could easily be developed that
would save operators both time and money without sacrificing the quality of the data or
diminishing the overall goal of the GHG emissions reporting rule under 40 CFR 98. It is IPAMS
understanding that this substantial dataset will be made available by EPA to state regulatory
agencies and the general public, and that statistically valid emission factors could be generated
from these data. IPAMS has worked recently in conjunction with the Western Regional Air
Partnership (WRAP) in a similar process that has developed improved criteria pollutant
emissions inventory data for the oil and gas industry in the Intermountain West. Accordingly,
IPAMS requests that EPA provide an option to allow reporters to use emission factors developed
from data gathered from the initial reporting year for subsequent reporting years.
Response: Upon further analysis and review, EPA has determined that certain sources could take
advantage of data collected in the first year and subsequent years to create company-specific
emission factors that can be applied to similar activities and are periodically updated. Todays
final rule includes several sources that apply this methodology, among which are gas well
drilling completions using hydraulic fracturing, gas well liquids unloading, and different
operational modes of large compressors (operating, standby pressurized and shut-down
depressurized). EPA determined that the activity data collected, in conjunction with company
and facility derived emission factors would provide cost-effective and suitable data for informing
future policy. EPA may consider new data as it becomes publicly available, such as from the
Western Regional Air Partnership, in future technical modifications of the rule.
13.1.1
POPULATION FACTORS

Delay reporting from the distribution sources for which the use of population emission factors
has been proposed until more accurate emission factors are available, or commit to updating
these emission factors as soon as more accurate ones are available;
Response: Please see the response to EPA-HQ-OAR-2009-0923-1065-11.
Sections 98.233(q) and (r) Fugitive Emissions
1307
Onshore Production Source Category: Fugitive emissions from valves, connectors, open ended
lines, pressure relief valves, compressor starter gas vents, pumps, flanges, and other fugitive
sources (such as instruments, loading arms, pressure relief valves, stuffing boxes, compressor
seals, dump lever arms, and breather caps for crude services).
API believes that the methodology for arriving at a component count must be significantly
changed to enable a reasonable approach to arrive at a component count for the hundreds of
thousands of equipment and sites subject to the proposed rule. As currently structured, the rule
would require a physical component count (along with concurrently gathering other information)
at each well-site and associated facility. APIs time and cost estimate for this inventory arrives at
some 677 man years of time (at 2,000 hours per man year) and $168.7 MM necessary simply to
inventory the specified components. To address this problem, API suggests the following:
- In lieu of the use of population counts and emission factors, API requests the option to use
alternative emissions calculations for fugitive components where the methodologies are federally
enforceable through state or federal permits.
- In a particular Sub-basin entity, operators tend to use similar equipment packages, such as
separators, compressors, dehydrators, and tanks, and similar site layouts or configurations. API
suggests that EPA modify the methodology for arriving at a component count for a particular
Sub-basin entity as follows:
1. Require a component count for a small number (5) of each type of major equipment package.
Inventory the number of each type of major equipment packages deployed in the Sub-basin
entity. Multiply the component counts by the number of major equipment packages to arrive at a
total component count for the major equipment packages.
2. Require a component count for a small number (5) of each type of site (e.g. single well-site,
multi-well site, tank battery, central production site, etc) for components which are not part of
the major equipment packages. Multiply the site component counts by the number of sites of
each particular type to arrive at a total component count for components not part of the major
equipment packages.
3. Sum the major equipment and site component counts for each Sub-basin entity to arrive at the
total component count for the Sub-basin entity and then apply the methodology specified in the
rule (Section 98.233 (r)) to arrive at a fugitive emission estimate for the Sub-basin entity.
4. Sum the fugitive emission estimates for all Sub-basin entities within an identified basin for
purposes of annual reporting at the Basin entity level.
By making these modifications to estimate component counts, EPA can achieve a more
appropriate balance between the burden of the rule and the amount of emissions covered while
still yielding good quality emission estimates. Adopting this methodology would also partially
address the issue of the industrys ability to physically meet the rule requirements in the time
contemplated by EPA.
1308
In addition, the minimum component size for leak survey should be defined (i.e., line size). API
suggests that components associated with line size or tubing less than 2-inch outside diameter
should be excluded because these components are expected to have minimal leakage while
significantly adding to the burden on industry for collection of data associated with this smallest
category of components. Ultimately, the leak detection and component count required should be
consistent with existing leak detection and repair (LDAR) regulations to minimize additional
burden required to collect data on these components.
Response: EPA disagrees with the commenters request to allow onshore production reporters
to count a small number of components and then to apply these counts across a sub-basin
entity to calculate equipment leak emissions, as this would not represent actual emissions.
EPA disagrees with the onshore facility definition as a sub-basin; please see response to
comment EPA-HQ-OAR-2009-0923-1305-46. In order to manage burden while still gathering
data necessary to inform future policy, EPA determined that, for equipment leaks in onshore
production, the proposed rules method to count the individual components and use population
based emissions factors can be replaced with major equipment counts. The reporter would only
count large equipment and apply a default average small component count based on the large
equipment count. This will result in significantly less burden to reporters than counting each
component (valve, flange, open-ended line, etc.) while sustaining the necessary quality of data,
and thus EPA has revised todays final rule. For more information, please see Equipment-Level
Population Emission Factors for Onshore Production, EPA-HQ-OAR-2009-0923, and Section
II.F of the preamble to todays final rule. However, reporters are still allowed the option to count
components individually for a facility and apply the appropriate population emission factors. For
counting components individually, EPA determined that tubing systems that are less than 0.5
inches in diameter are a small emission source and therefore, neednt be reported; EPA has
revised todays final rule accordingly.

Section 98.233(r) Population count and emission factors
El Paso does not track all fugitive component counts for its pipeline and onshore oil and gas
production operations. In evaluating the impact of the proposed rule requirements, El Paso
estimates that it would have to locate and log over 1 million fugitive components across
operations in ten states. Given the sheer number and geographic dispersion of fugitive
components associated with onshore oil and gas operations alone, it is not possible to definitively
account for all components, nor is it possible for EPA to verify the accuracy of the reported
component counts for all operators. El Paso strongly encourages EPA to consider the trade off in
emission accuracy between the use of component level emission factors applied to a count of
components with a high degree of uncertainty due to the vast number of sources versus a slightly
less accurate equipment level emission factor applied to a count of equipment that is known to a
high degree of certainty. Furthermore this approach is more easily verified by EPA as well as the
1309
operators.
Given the geographic dispersion and number of small fugitive components, in addition to the
disproportionate cost relative to the quantity of emissions, the rule should provide equipment
level population-based counts and emission factors from industry accepted protocols such as
the API Compendium, INGAA guidelines, etc. to quantify population-based fugitive
emissions.
EPA states in the Preamble that the purpose of the Mandatory Reporting Rule is to guide policy
development for future permitting and compliance purposes. General equipment-level emission
counts are acceptable for this purpose, and therefore should be acceptable for petroleum and
natural gas systems.
Response: EPA agrees with using onshore production major equipment counts for equipment
leaks. Please see the response to comment EPA-HQ-OAR-2009-0923-1151-126.
Commenter: Kim Dang
Use of Population Emission Factors for Onshore Petroleum and Natural Gas Production and
Gathering Pipelines. Although Kinder Morgan appreciates EPAs emphasis on using population
emission factors where possible, EPA has greatly underestimated the burden associated with
applying population emission factors to the onshore petroleum and natural gas production sector.
Many of the countrys approximately 450,000 wellheads, as well as the equipment associated
with those facilities, do not have existing component counts. Carrying out component surveys at
these sites would require considerable time and would be a highly labor-intensive process.
Accordingly, Kinder Morgan recommends that EPA allow facility-level emission factors to be
applied to onshore petroleum and natural gas production facilities, rather than requiring the
application of component-specific emission factors. This method will yield data with acceptable
accuracy and at much more reasonable cost than the proposed requirement.
Consistent with our recommendation above that Subpart W not require reporting of emissions
from gathering pipeline segments, Kinder Morgan also recommends that the list of components
for which population emission factors would be required under proposed 40 C.F.R. SECTION
98.233(r) be revised to exclude gathering pipeline fugitives in the onshore petroleum and natural
gas production sector 189 and the natural gas processing sector. 190
Response: EPA has revised t todays final rule for equipment leaks in onshore production and
allows the option to count major equipment and apply default average component counts per
primary equipment; see the response to EPA-HQ-OAR-2009-0923-1151-126. EPA considered
189
See also proposed 40 C.F.R. SECTION 98.232(c)(9).
190
See also proposed 40 C.F.R. SECTION 98.232(d)(8).

1310
the comment to use facility level emission factors for onshore production, and has decided that it
is not appropriate for the rule, as facility level emission factors are not available for the onshore
production facility defined in this rule. As explained in Section II.E of the preamble to todays
final rule, EPA has decided not to include monitoring and reporting requirements for gathering
lines and boosting stations at this time.

Proposed Rule Section 98.233(r) Population count and emission factors. El Paso recommends
the use of a facility emission factor to be used as an alternative to population count for standard
components at processing plants, onshore petroleum, transmission compression, underground
storage, and LNG import facilities.
Response: EPA disagrees with the comment that facility level average emission factors should
be used for gas processing facilities, onshore production, transmission compression, underground
storage, or LNG import and export facilities. With regards to onshore production facility level
emission factors, please see the response to EPA-HQ-OAR-2009-0923-1024-27. For gas
processing facilities, transmission compression, underground storage, or LNG import and export
facilities the facility level average emission factors are developed using typical facility activity
factors and aggregated component emission measurements, and would not be reflective of the
reporters emissions, but rather average industry emissions. EPA deems this unacceptable data
quality for the purposes of informing future policy. Please see Section II.L of the preamble to
The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98). EPA is allowing
equipment level reporting for onshore production; please see the response to EPA-HQ-OAR
2009-0923-1151-126.

Cost of Component Inventory. The discussion in the proposed rule demonstrates that there is a
serious need on EPAs part to better understand natural gas distribution and transmission
systems. EPAs proposal to use a population count of each fugitive emissions source (e.g., source
types such as valves, connectors, etc.) multiplied by a population emissions factor simply will
not work. Unfortunately an inventory of the types and numbers of components that are required
to be inventoried for compliance with the proposed regulations is not readily available and it
would be very costly and time consuming to compile this information.
1311
Response: EPA disagrees with the comment on compliance cost. Please see the response to
comment EPA-HQ-OAR-2009-0923-1020-5. EPA has analyzed the use of population emission
factors and determined that it is a less burdensome method, and this remains in todays final rule.
Please see the Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background Technical Support Document (TSD) found in docket (EPA-HQ-OAR-2009-0923
0027) Section Monitoring Method Options for further information.

The section on population count and emission factors includes emissions from gathering
pipelines in Section 98.232(d)(8). This will require an operator to count and track every piping
components [sic] along tens of thousands of miles of highly dispersed gathering pipeline,
including tens of thousands or meter sites and valve settings. EPA has greatly underestimated the
costs and resource requirement to conduct this population count, or they have perhaps failed to
even consider the need to do this population count. Additionally, the gas quality can vary at
every connection to a gathering pipeline. This would require that each pipeline segment be
counted and tracked separately with the associated gas analysis applied to each segment.
Response: EPA has decided not to include monitoring and reporting requirements for gathering
lines and boosting stations at this time. Please see response to comment EPA-HQ-OAR-2009
0923-1024-27.
Restricting component population counts to components that are above a minimum size
threshold.
Response: EPA agrees that a minimum size threshold is necessary. Please see the response to
EPA-HQ-OAR-2009-0923-1151-126.

Clarify Component Count Requirement .Proposed section 98.236(c)(19) Data Reporting
Requirements states: For fugitive emission sources using emission factors for estimating
emissions report the following: (i) Component count for each fugitive emission source. If
1312
reporting will be required by component count instead of by station or facility, AGA

recommends that EPA:
-
Clarify that only a count of leaking components by type are required to be reported, not a
count of all components by type; and
Establish a size threshold for components, such as 2 nominal diameter.
Response: EPA has clarified the data reporting requirements for equipment leak sources that
use emission factors. Reporters that use leaker emission factors must report the total count of
each type of component found leaking in a survey, as the commenter suggested. In addition,
reporters that use population emission factors must report the total number of each type of
component. EPA agrees, and a size threshold of 0.5 inches has been established. Please refer to
EPA-HQ-OAR-2009-0923-1151-126 and EPA-HQ-OAR-2009-0923-1152-8.
98.233(r) references to 98.232(e)(6) and (f)(4) should be deleted: Low bleed pneumatic
devices are identified as a source for gas transmission compression in 98.232(e)(6) and the
emission estimation method is appropriately detailed in 98.233(b). This estimate uses
population-based EFs from Table W-3. However, 98.233(r) also indicates that the estimation
method in that paragraph applies to 98.232(e)(6). Since 98.233(b) already addresses estimates
for this source type, any further delineation in 98.233(r) would be duplicative. It is apparent that
the reference to (e)(6) in 98.233(r) is an error and (e)(6) should be deleted from the list of
emission sources in 98.233(r).
In addition, the reference to (f)(4) as an emission source in 98.233(r) is inappropriate and
should be deleted. Similar to the discussion above, 98.232(f)(4) identifies low bleed pneumatic
devices as a source for underground storage and the estimation method is already addressed
under 98.233(b).
Response: EPA has reviewed this comment and has decided to remove the reference to natural
gas pneumatic device venting from 40 CFR 98.233 (r) in todays final rule, as those sources have
their own specific methodologies..

High Bleed and Low Bleed Natural Gas-Driven Pneumatic Devices 98.233(a) and (b) provide
emission estimates based on emission factors for high bleed and low bleed natural gas-driven
pneumatic devices, respectively. Low bleed devices are defined as devices with gas bleed rates
1313
less than or equal to 6 scf/hr. For low bleed devices, a Subpart W population emission factor is
used. Emissions from all low bleed devices are calculated from the count of low bleed devices,
the population emission factor, and an assumption of 8,760 operating hours per year. High bleed
devices are defined as devices with gas bleed rates greater than 6 scf/hr. Equation W-1 is used to
estimate gas emissions from each high-bleed device: [See original attachment for equation]
The emission factor (Bs) is the device bleed rate provided by the manufacturer or based on data
for a similar device. Annual operating minutes (T) for each device must be determined.
INGAA accepts these methods for pneumatic device emission estimates with the request that two
clarifications be provided in the Final Rule:
(1) time basis for high bleed devices is facility operating time and not related to time
based on device actuation; and
(2) definitions of low-bleed pneumatic device and high-bleed pneumatic device in
98.6 are revised to clarify that applicability is based on continuous bleed rate and
facility operating time.
Further, it is INGAAs understanding that pneumatic devices addressed under
98.233(a) and (b) are continuous bleed devices and not intermittent devices. Thus,
valve actuators with intermittent releases are not covered by this section and
intermittent release valve actuators do not meet the definition of high-bleed or lowbleed pneumatic device under Subpart W. For natural gas transmission and storage,
valve actuator components would be screened during the leak survey required under
98.233(q).
Response: Todays final rule has been revised such that high bleed pneumatic devices will
report emissions using an emission factor with an assumption of 8,760 operating hours per year.
EPA determined that monitoring the operational time for each high-bleed pneumatic device is
overly burdensome. EPA agrees that the definition of pneumatic devices needed clarification,
and therefore in todays final rule has clarified the definition for high-bleed and low-bleed
pneumatic devices to state that they are continuous bleed devices. Furthermore, in todays final
rule, intermittent bleed devices are also required to report emissions. A definition for
intermittent bleed devices has been added to 40 CFR 98.6. In addition, this commenter
misinterpreted the EPAs intent regarding leak detection on valve actuator components in
transmission and storage facilities. EPA does not require reporters to conduct leak detection
surveys on pneumatic devices in the transmission and storage sector.
Commenter:
First, the term definitions for Eq. W-19 refer to "facilities listed in 98.230(b)(3) through (b)(8)."
However, 98.230(b) is marked reserved, so this reference is incorrect.
Response: EPA agrees and in todays final rule, the reference has been revised to refer to the
appropriate facilities.
1314

Emission Factors. For LDC pipelines and below grade M&R stations, the proposed rule requires
a population count of emissions sources and use of population emissions factor to estimate
fugitive emissions. While the Clean Energy Group agrees with this approach, we recognize that
fugitive and vented emission factors for the oil and gas sector are outdated and uncertain, and
that emission factor studies are ongoing that will likely provide more accurate emission factors
for the industry. Therefore, the Clean Energy Group recommends that EPA delay requiring
emissions reporting from these sources until more accurate emission factors are available.
Alternatively, we encourage EPA to commit to updating these emission factors as soon as more
accurate ones are available.
Response: EPA disagrees with this comment to delay reporting from sources that calculate
emissions using population emission factors. EPA used best available, public data to develop the
emission factors in Table W-1 through W-7 of todays final rule. If and when new data become
available, EPA will consider the need to update the emissions factors. All other alternatives
considered were either less recent or more prone to erroneous calculations than the data EPA
used. Moreover, EPA developed the emission factors with the intent of minimizing the reporting
burden on reporters. Where used in todays final rule, population emission factors are the least
burdensome method for reporters, while sustaining the necessary quality of data. In addition,
EPA will evaluate the natural gas distribution data received through the MRR, which are based
on emission factors and methodologies in subpart W. EPA will also evaluate other emission
factors that EPA may receive. EPA may also consider using direct measurement of equipment
leaks and vents if EPA deems the natural gas distribution emissions data received through the
MRR is insufficient to inform policy.

If distribution facilities must report fugitive emissions from above-ground M&R stations, allow
the use of population emission factors in order to reduce reporting costs.
If the use of population emission factors to report emissions from above-ground M&R stations is
not allowed, redefine distribution facilities to exclude small, remote sub-systems in order to
reduce reporting costs.
Response: The commenter made several assumptions which were not consistent with EPAs
intent. One was the assumption that EPA required leak detection and calculating emissions from
customer meters. Reporters are not required to calculate emissions from customer meters. EPA
1315
has clarified this in todays final rule. Please see Section II.F of the preamble to todays final
rule. Furthermore, in todays final rule, above-ground meters and regulators at city gate station
at custody transfer are required to use leak detection and leaker emission factors to calculate
emissions. For above-grade meters and regulators at city gate stations not at custody transfer,
reporters must develop a facility emission factor using emissions and meter count from the leak
detection surveys conducted at the custody transfer city gate stations. This facility emission
factor, along with a count of meters at the non-custody transfer city gate station, is used to
calculate emissions from above-grade meters and regulators at city gate stations not at custody
transfer.
Monitoring methods and emissions quantification for: LDC pipeline mains and service lines
(98.232(i)(3) and (4))
How do you calculate fugitives from distribution pipelines using the population count method?
The emission factor in Table W-7 says the units are scf per hour per service. What does EPA
mean by "service"? Did EPA intend to mean miles of pipeline? If EPA means miles of pipeline,
the only way to reduce emissions would be to change the pipe material. It would seem that if a
company where able to show proper maintenance records than the factors should be reduced.
Response: It appears as though the commenter misunderstood the method of calculating
emissions from LDC pipeline mains and service lines. To calculate emissions from distribution
mains, the number of miles of distribution mains pipeline are multiplied by a population
emission factor with units of standard cubic feet per hour per mile of pipeline. To calculate
emissions from distribution service lines, the number of service lines branching off from the
main distribution line is multiplied by a population emission factor with units of standard cubic
feet per hour per number of services. EPA does not require the use of leak detection and leaker
emission factors, which would better represent actual emissions, because this would be unduly
burdensome.
13.1.2
LEAKER FACTORS

Third, EPA should delete the requirement to apply leaking emission factors to leaking
components at city gates and above ground M&R stations, and instead, EPA should allow LDCs
to use a facility level emission factor for city gates and both above ground and below ground
M&R stations.
1316
Response: EPA disagrees with using a facility level emission factor for above and below ground
meters and regulators at city gate stations. Please refer to EPA-HQ-OAR-2009-0923-1065-4.

Cost of Improving Emission Factors
It is well known that the accuracy of existing emission factors varies by as much as 2,000%.
There have been significant efforts made to improve those accuracies within the last two years
by the AGA and individual companies, such as Southwest, who have implemented direct
measurement programs to further enhance the base information from which emission factors are
derived. While any new information will be helpful, the cost of this work and the time needed to
do the work does not appear to be adequately addressed in the justification for the proposed rule.
EPA found that emission factors generated from the Clearstone studies related better to methanerich stream fugitives and were more appropriate than other emission factors developed for highly
regulated refinery and petrochemical plants on VOC emissions. Therefore, EPA is using
emissions data from the Clearstone studies as the basis for the leak factors proposed in this rule.
Due to the extremely short comment period, Southwest does not have time to thoroughly
evaluate the emission factors from the Clearstone studies. Southwest has no comfort that this
alternative approach better estimates annual facility emissions without resulting in additional
reporting burden for the facilities. Multiple surveys only add to the confusion that EPA has
already created in attempting to promulgate this rule.
Response: EPA disagrees that the emission factors in the rule are inaccurate. EPA used the
Clearstone studies to develop the emission factors in todays final rule, as it is the best available
public data on transmission and distribution facilities. Please see the TSD EPA-HQ-OAR-2009
0923. All other alternatives considered were either less recent or more prone to erroneous
calculations than the data EPA used. Moreover, EPA developed the emission factors with the
intent of minimizing the reporting burden on reporters. Where used in todays final rule,
emission factors are the least burdensome method for reporters that sustains the necessary quality
of data.
EPA disagrees that multiple surveys are confusing. Multiple surveys would improve data
quality, as a leak would not be assumed to be ongoing based on the data from the last regular
leak survey. Therefore EPA is allowing reporters to voluntarily submit data from multiple leak
detection surveys in todays final rule. EPA deliberated on the issue and determined that
assuming that a leak perpetuates for the duration of a reporting period may overestimate
emissions. Allowing multiple surveys allows reporters to adjust emissions for a component
found leaking and not leaking, respectively, between two consecutive leak detection surveys.
EPA disagrees with the comment regarding the benefits of todays final rule. Please see the
Section III.E of the preamble to todays final rule. In addition, EPA does not need to include in
1317
the rules cost burden analysis the cost incurred by reporters to conduct their own emission
measurement studies that are outside the scope of the rule.

Measurement Complexities
Although simple in principle, the proposed measurement techniques require substantial
experience to perform correctly. Considerations include:
1. H-Flow enclosures must be placed so as not to restrict air flow and allow full capture of
leaks.
2. The back ground probe for the H-Flow sampler must be placed in the correct position to
determine if there is any interference from other sources as well as to determine whether
or not complete capture of the leak has been achieved.
3. The vent bag technique must be first practiced under metered conditions in order to be
confident of the results. Personnel must be aware of how to eliminate potential errors due
to leaks, restricted openings, or flow induced by Venturi effects.
4. Some vents may be manifolded together, sometimes going to a larger diameter (six feet)
silencer. Larger diameter vent stacks have been measured previously with the use of
covers, but this technique requires practice and great attention to safety considerations. It
may not be possible to know which source is the leak in these cases.
Contribution of Standard Components to Total Leakage
There may be an overemphasis on the importance of standard components at compressor stations
due to Clearstone Engineerings studies at gas plants in 2002 and 2006. Although these studies
were detailed and well executed, gas processing plants have a much larger ratio of standard
components per compressor unit than is found at transmission compressor stations. For instance,
in the PRCI study done at thirteen transmission compressor stations, the number of components
per compressor unit was a little over 400. The Clearstone studies dont list a complete inventory
of compressor units, but if there were twenty units at each site, the ratio of components to
compressor units would be approximately 1000, about two and one half times greater than at
transmission sites. Additionally, standard components at processing plants may be subjected to
greater heat and vibration in some areas than at transmission compressor stations.
The 2002 Clearstone study (EPA, 2002) indicated that connectors and valves accounted for over
50% of the total leakage (Figure 10 of that study), whereas at transmission stations that number
is more likely to be 20% (Howard et al, 1999). Additionally, the 2006 Clearstone study (KSU,
2006) indicates that connectors and valves were down to approximately 32% (Figure 13 of that
study). Open ended lines contributed another 30%, but this category apparently includes
compressor blow down valve leakage which would probably dominate the leakage.
It is understood that for the purposes of the proposed rule that different leaker emission factors
have been developed for transmission as opposed to processing. However, the key point is that
standard components in transmission dont contribute to the total leakage as much as the
1318
Clearstone studies might be interpreted to suggest.

Response: With regard to measurement complexities, EPA disagrees that the four points the
commenter discusses will prevent compliance with subpart W. In todays final rule, leak
measurement using the instruments listed is only required for compressors. Reporters are to
measure the manifolded vents in the operating mode in which the compressor is found. The rule
provides guidance on how to properly use calibrated vent bags and high volume samplers;
however, it also states that it is the responsibility of the reporter to follow all manufacturer
instructions to ensure quality measurement. Reporters may also contract these measurements out
to trained personnel. Todays final rule further allows reporters to apply for the use of best
available monitoring methods, should the reporter have trouble accessing staff to perform this
work. Please see Section II.F of the preamble to todays final rule. In addition, the use of
acoustic detectors is also allowed in todays final rule for through valve leakage for compressors
and transmission storage tanks. EPAs decision to include standard components is based on the
criteria under Selection of Emissions Sources for Reporting in the Greenhouse Gas Emissions
from the Petroleum and Natural Gas Industry: Background TSD (EPA-HQ-OAR-2009-0923
0027). The standard components met the criteria and were subsequently required to report in
todays final rule. Upon further analysis and review, EPA has revised the leaker emission factors
for transmission and processing to be the same. Please see the memo Revisions to Processing
Leaker Emission Factors in Rule Table W-2 in EPA-HQ-OAR-2009-0923.

Detection/Leaker Emission Factors for Standard Components at Transmission Compressor
Stations.
Currently, Section 98.233 (q) requires surveying standard components (such as connectors and
valves, as opposed to vented components) with an IR camera and then applying a leaker
emission factor to the number of leaks found at each component type.
Its clear that EPAs intent was to reduce labor costs in this area. However, this approach still
requires a substantial amount of effort that does not provide a large improvement in the accuracy
of the reported emissions. It would also distract from collecting accurate data from the vented
components that make up the majority of fugitive emissions.
The contribution of leakage from standard components averages about 20% of the total leakage
at transmission compressor stations (Howard et al., 1999 this was a joint study sponsored by
PRCI, GRI, and EPA.). Because the proposed method still uses an emission factor as opposed to
a measurement, there will still be uncertainty in this method. The EPA Natural Gas Star program
has previously developed emission factors for compressor stations based on the number of
compressor units at the site (EPA, 2000). These factors include both standard and vented
components but could easily be recalculated to assess only standard components.
1319
Although the proposed method of using leak detection combined with leaker emission factors
would be expected to give better results than applying a facility emission factor, this
improvement only applies to approximately 20% of the total facility leak rate. Consequently, the
improvement in the reporting accuracy for the entire facility is probably less than 5%, while the
effort involved might require as much as 50% of the total leak survey effort.
Allowing a facility emission factor for standard components as an option to leak screening lets
facilities focus on making measurements at the vented components that account for 80% of the
total emissions. This full focus should improve the accuracy of the data collected from these key
sources.
Should EPA still feel that leak surveys of standard components are essential, it would be better to
allow more options than just the IR camera to conduct these surveys. Although these cameras are
useful tools, screening with flame ionization detectors, catalytic oxidation/thermal conductivity
detectors, and soap solutions have been proven to be effective (Howard et al., 1999; EPA, 2002;
KSU, 2006) and should not be excluded.
Its possible that EPA felt that the IR cameras would be the most cost effective approach, and for
some companies, that may be true. However, other companies might want to have in-house
personnel conduct leak screening with techniques that they already know how to use.
Clearstones 2006 study (KSU, 2006) notes that a 2 person crew with an IR camera can survey
approximately 6400 components per day. The PRCI study (p. A-1-14) found that experienced
personnel could survey transmission station components using a combination of soap and
catalytic oxidation/thermal conductivity detectors at a rate of 350 components per person per
hour, or 5600 components per two person crew per eight hour day. During this time study, the
personnel found 99.6% of combined leaks found by using this method and a Foxboro TVA
1000, so these methods are clearly still effective.
Facilities that have access to IR cameras and trained personnel could apply the IR camera if they
so desired. Other facilities without these resources could rely on past experience and conduct the
leak screening on a flexible time schedule without necessarily acquiring more equipment or
hiring contractors.
A further discussion of the contribution of standard components is given in the appendix of this
letter.
Response: EPA disagrees, and did not include transmission compressor station facility-level
average emission factors because they do not provide sufficient quality emissions data to inform
future policy; please refer to EPA-HQ-OAR-2009-0923-1011-19. EPA selected leak detection
and leaker emissions factors versus other options as it provides a more accurate calculation of
emissions from each piece of equipment and reduces burden. Please see Section II.E of the
preamble to the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002).
Further, EPA included transmission compressor station equipment leaks as they fell within the
80/20 criteria in the TSD. Please see the TSD in EPA-HQ-OAR-2009-0923. EPA has noted the
prevalence of other methodologies for leak detection and has included several additional leak
1320
detection techniques including flame ionization detectors, catalytic oxidation/thermal

conductivity detectors, and soap solutions as per Method 21; please see Section II, Petroleum and
Natural Gas Systems, of the Preamble for further details.

Inclusion in the MMR: We think it is important to take into consideration the fact that both
industry and governments continue to conduct field research to improve the quality and basis for
fugitive emission factors. We know that these factors will continue to be updated, possibly quite
frequently. We strongly suggest that EPA remove the emission factors from the rule itself to a
reference source where emission revisions can be more readily published.
Use of GRI-EPA emission leaker factors: The EPA's TSD (p. 40) cites several concerns with
these emission factors concluding that ''this method for estimation of the emissions is not
considered to be appropriate for a mandatory GHG reporting program." Additionally, these
factors do not align with more recent and representative field work done in Canada. The use of
these default emission factors risks generating a skewed national-level inventory for the sector
that is not representative of current-day LDC facilities. It is also worth noting that odorized
systems, which are the norm in LDC service, have demonstrably lower leak emission rates than
non-odorized systems.
Response: EPA disagrees that emission factors in the rule are not representative of the U.S.
industry. EPA used best available, public data to develop the emission factors in Table W-1
through W-7 of todays Final Rule. In addition to GRI studies, for distribution EPA also used
data from Canadian studies Measurement of Natural Gas Emissions from the Canadian Natural
Gas Transmission and Distribution Industry 2007 and Handbook for Estimating Methane
Emissions from Canadian Natural Gas Systems 1998. Please see the TSD in EPA-HQ-OAR
2009-0923. All other alternatives considered were either less recent or more prone to erroneous
calculations than the data EPA used. Moreover, EPA developed the emission factors with the
intent of minimizing the reporting burden on reporters while maintaining the necessary quality of
data. EPA disagrees that EPA should remove the emission factors from subpart W and place
them in another reference technical document. Please refer to EPA-HQ-OAR-2009-0923-1299
5. In addition, EPA will evaluate the natural gas distribution data received through the MRR,
which are based on emission factors and methodologies in subpart W. EPA will also evaluate
other emission factors that EPA may receive. EPA may also consider using direct measurement
of equipment leaks and vents if EPA deems the natural gas distribution emissions data received
through the MRR insufficient to inform policy.

Organization: Verdeo Group
Commenter: John Savage
1321

S98.233 (q) Leak detection and leaker emission factors
This section requires the use of component counts and emission factors for calculation of
emissions detected from specified sources. A more accurate approach would be the use of a
measurement device such as a high volume sampler or comparable methane leak detector.
Verdeo recommends that Section (q) be amended to allow (not require) companies the option of
measuring leaks with such devices. Some companies may already be using such equipment as
part of their site survey to collect data for the MRR. They may elect to invest the incremental
time required to take an accurate direct measurement of leak rates in order to help them prioritize
future investments in leak reduction processes or technologies. If they choose to do so and have
the data available, they should be encouraged to submit it in their report. The company should
simply indicate on their report if the leak data was derived from a direct measurement or the use
of an emission factor.
Response: EPA disagrees that it should provide an option for reporters to submit measured data
for equipment leak emissions. In the April 2009 proposed rule, EPA required reporters to
measure their equipment leaks; however, many potential reporters commented that the cost
burden was too high. EPA accepted these comments and revised the rule to population emission
factors and leaker factors to reduce compliance burden on industry while maintaining the
necessary quality of data to inform policy. Please see Section III of the preamble to the April
2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002).
Commenter:
Natural gas transmission and distribution facilities are highly regulated by other regulatory
bodies, including the U.S. Department of Transportation through its Pipeline and Hazardous
Materials Safety Administration, the Federal Energy Regulatory Commission, and state
regulatory agencies. While these regulatory bodies do not specifically regulate greenhouse gas
emissions, they are responsible for developing and enforcing strict gas leak response and repair
requirements and safety standards. As a result of the imposition of these requirements,
MidAmerican believes that on both the natural gas transmission and distribution side, EPAs
proposed fugitive default factors are overstated for leaks; that the leak duration assumptions (365
days) dramatically overstate the true duration of a leak and, as a result, overstate emissions; and,
that the requirement to annually measure leaks from regulator stations is unnecessary.
MidAmerican recommends that EPA obtain additional information regarding leak detection,
response, and repair be sought from the pipeline safety organizations prior to finalizing the
monitoring and reporting rules.
Response: EPA disagrees that rule is unnecessary because of DOT regulations. EPA conducted
extensive research on all applicable transmission and distribution-sector related regulations in
1322
place today that might have similar requirements to subpart W, including consultations with
Department of Transportation experts. Please see the rulemaking docket (EPA-HQ-OAR-2009
0923) Understanding the Substance of the DOT Regulations and Comparing Them to the
Subpart W Requirements. In todays final rule, EPA does not require leak detection at all
regulator station; please see the response to EPA-HQ-OAR-2009-0923-1065-4. In addition, leak
detection is not required at below-grade M&R, and instead use population emission factors.
EPA agrees that assuming a leak duration of 365 days may overestimate emissions and is
allowing adjustment of emission in todays final rule; please see the response to EPA-HQ-OAR
2009-0923-1014-9.

The Final Rule should allow LDC operators to estimate emissions at above-ground M and R
Stations together with the use of recent emissions data.
By operation of Section 98.233 (q), all LDC facilities identified by Section 98.232(i)(1) are
required to be leak surveyed annually. Yet, as noted, under PHMSAs pipeline safety regulations
LDCs conduct leakage surveys of their entire system at least once every 5 years. Some LDC
operators leak survey their above ground facilities at least once every three years to coincide
with a 3 year requirement to inspect the facilities for atmospheric corrosion.
APGA urges EPA to allow LDCs the option of maintaining their existing 3 or 5 year leakage
survey cycles and to allow an estimate of emissions using an emission factor for facilities
identified in Section 98.232 (i) (1) in years when no leakage survey is conducted on a given
facility. A population emission factor would need to be developed.
Response: EPA disagrees, as current leakage surveys as prescribed by DOT regulations are not
applicable to subpart W as they focus on safety and therefore gas concentration levels rather than
specific emissions sources and the volume of emissions from said leak source. Therefore EPA
will not allow LDCs to apply in subpart W the data gathered to meet the safety requirements
under existing DOT regulations. Please see response to comment EPA-HQ-OAR-2009-0923
1082-7. EPA disagrees that direct measurement of sources is not necessary. Please see Section
II.L of the preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part
98). EPA disagrees with less than annual reporting. Please see Section II.H of the preamble to
The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98).

1323

(98.233 (q)) What is the definition of a for use in the fugitive emission equation (Eq. W-18)
Response: In the rule paragraph 98.233(q), the definitions of emission sources are found in
subpart W and in the Final MRR.
XVI. Proposed Rule Section 98.233(q) Leak detection and leaker emission factors El Paso
requests monitoring flexibility and urges the use of a facility emission factor to be used as
an alternative to Infra-red (IR) camera leak detection/leaker emission factors for standard
components at transmission compression and underground storage facilities.
Response: EPA disagrees, and did not include facility level average emission factors because it
does not provide sufficient activity data to inform future policy; please refer to EPA-HQ-OAR
2009-0923-1011-19.

VIII. Proposed Rule Section 98.233(a) Natural gas pneumatic high bleed device venting.El
Paso encourages the EPA to provide a default leak rate for those devices where
manufacturer and model is not known.
Response: Based on comments received, EPA has revised todays final rule to allow the use of
population emission factors for pneumatic device venting emissions, this will reduce burden on
the industry while maintaining the necessary quality of data to inform policy. This will eliminate
the need to determine the device manufacturer model. EPAs decision to include a default leak
rate was partially related to the difficulty of ascertaining the manufacturer and model in certain
cases. EPA determined that the burden of determining the manufacturer and bleed rate of each
reporting pneumatic device was unnecessary, as the collection of activity data and the application
of emissions factors will not adversely affect the quality of reported data allowing consideration
of future policy options.

1324
General Comments Use of Leak Detection and Leaking Component Emission Factors
98.233(q) requires direct measurement of leaks. In the preamble on page 18623, the EPA
concedes that fugitive emissions detection for onshore petroleum and natural gas production is
particularly challenging due to large geographical operations. PAW agrees with this statement.
Also on page 18623 of the preamble, the EPA proposes that, if a component fugitive emission is
detected, emissions are assumed to occur the entire 365 days in the year. What if a leak is
repaired the same day it is detected? It seems the EPA should at least let the operator prorate the
leak from the beginning of the year to the day it was known to be repaired.
Response: EPA agrees to allow multiple leak detection surveys, please see the response to EPA
HQ-OAR-2009-0923-1014-9.

Allow the Use of a Leak Longevity Factor to Permit More Accurate Estimates of Methane Leaks
at M&R and Gate Stations
EPA proposes to apply a leaker emission factor to components found to be leaking during leak
surveys of M&R and city gate stations. These leaker emission factors assume that the leak occurs
for 365 days during the year, which will significantly overstate emissions for most leaks. AGA
recommends an alternative that will lead to more accurate emissions estimates by including a
leak longevity factor in the calculation.
LDCs typically maintain systems that allows them to estimate the magnitude of a methane leak
caused by a contractor hitting a gas line, in order to facilitate billing the contractor afterward for
the resulting third party damage and the costs of repairing the leak. In addition, LDCs respond
to leak calls generally within two hours or less, they know the leak began shortly before the
caller was able to smell the odorized gas, and the LDC fixes the leak promptly. Thus for such
leaks, it is possible to put a time bracket on the leak of 6 hours or less. To estimate emissions
from the leak, one can apply the allocable portion of the 365-day leaker emission factor (e.g. a
leak lasting one day would equate to the leaker emission factor divided by 365.
Response: EPA disagrees with the comments proposed method for equipment leaks and leaker
factors for LDC, since it uses methods under DOT regulations. Please see response to comment
EPA-HQ-OAR-2009-0923-0955-7. However, EPA is allowing multiple leak detection surveys;
please see the response to EPA-HQ-OAR-2009-0923-1014-9. The rules multiple leak detection
surveys method calls for the component found to be leaking to be calculated from the beginning
of the year or since the previous leak detection survey.
Organization: Sempra LNG
Commenter: David M. Cobb
1325

Under 98.233 (q), the EPA assumes that, if a fugitive emission source is detected, the emissions
would occur the entire 365 days of the year. This assumption is inaccurate, as LNG facilities
cannot and do not permit such conditions as LNG valve leakage to continue after detection.
When leaks are detected, they are quickly corrected.
Response: EPA is allowing multiple leak detection surveys; please see the response to EPA-HQ
OAR-2009-0923-1014-9. The rules multiple leak detection surveys method calls for the
component found to be leaking to be calculated from the beginning of the year or since the
previous leak detection survey.

Commenter:
General Comments Use of Leak Detection and Leaking Component Emission Factors
98.233(q) requires direct measurement of leaks. In the preamble on page 18623, the EPA
concedes that fugitive emissions detection for onshore petroleum and natural gas production is
particularly challenging due to large geographical operations. Yates agrees with this statement.
Also on page 18623 of the preamble, the EPA proposes that, if a component fugitive emission is
detected, emissions are assumed to occur the entire 365 days in the year. What if a leak is
repaired the same day it is detected? It seems the EPA should at least let the operator prorate the
leak from the beginning of the year to the day it was known to be repaired.
Response: The commenter misinterpreted the requirements of the proposed rule for leak
detection and leaker emission factors. The EPA does not require reporters to directly measure
the leaks. Instead, EPA requires reporters to determine leaking components and apply a leaker
emission factor. EPA agrees with the commenters point about improving the accuracy of
emissions estimates to account for leak repairs and is allowing multiple leak detection surveys to
address this. Please see the response to EPA-HQ-OAR-2009-0923-1014-9.

Organization: Paiute Pipeline Company
Commenter: Jeff Maples
Revise and Clarify Provisions for LNG Storage Facilities
EPA should not apply leaker emission factors at LNG facilities because leaks are rare and
fixed immediately. EPA assumes that if a fugitive emission source is detected, the source must
have continued to leak for all 365 days of the year. This hypothetical assumption would vastly
overstate emissions and would not yield useful data for policy development. Leaks are rare at
1326
LNG facilities and located / resolved in a timely manner. These facilities are very tight by design
and operation. LNG facilities cannot and do not permit conditions such as LNG valve leakage to
continue after detection. Detection of leaks is rapid due to installed fixed gas detection, low
temperature spill detection and physical facility inspections by qualified operating staff. When
leaks are detected, they are generally corrected immediately. Thus, using a leaker emission
factor that assumes the leak continues all year would seriously overstate fugitive emissions at an
LNG storage facility or import terminal.
Response: EPA disagrees that LNG emission factors in the rule will not yield useful data to
inform future policy. EPA is allowing multiple leak detection surveys; please see the response to
EPA-HQ-OAR-2009-0923-1014-9. The rules multiple leak detection surveys method calls for
the component found to be leaking to be calculated from the beginning of the year or since the
previous leak detection survey.

Organization:
(98.233 (q)) The source category for Leak detection and leaker emission factors applies to
emission sources with streams with gas contents greater than 10 percent CH4 plus CO2 by
weight. We propose this source category be revised to include only CH4, as the infrared
absorption curve of CO2 does not allow CO2 to be visible using current Optical Gas Imaging
technology. Current Optical Gas Imaging cameras can locate gases with absorbance curves in the
3.3 to 3.5 micrometer range of the electromagnetic spectrum, while the absorbance curve of CO2
is in the 4.17 to 4.5 micrometer range of the electromagnetic spectrum and therefore cannot be
easily detected using OGI technology.
Response: EPA recognizes that the infrared cameras commercially available today do not detect
CO2. EPA considered the sources of GHG equipment leak emissions and determined that with
rare exceptions (e.g., CO2 EOR) the gas streams are predominantly hydrocarbons detectible by
IR camera and typically have a minor CO2 content. The method referenced for use of the IR
camera requires the camera to be able to detect the potentially leaking constituents at a certain
leak rate, and this will insure that the IR camera is a suitable leak detection method for streams
required to be surveyed in todays rule.

Commenter:
Source Type: Fugitives from valves, connectors, OELs, pressure relief valves, meters, and
centrifugal compressor dry seals, pumps, flanges, starter gas vents and other fugitive sources.
Regulatory reference for calculation/ monitoring requirements: 40 CFR 98.233(q)
1327
Total number of this type of emission source found to be leaking
For onshore natural gas processing facilities, concentration of GHGi, CH4 or CO2, in the
total hydrocarbon of the feed natural gas; for other facilities listed in 98.230(a)(3)
through (a)(8), GHGi equals 1.
Total time the specific source associated with the fugitive emission was operational in the
Comment: Direct measurement of leaks is very burdensome and expensive, particularly since
many upstream activities requiring FLIR surveys are operated in remote locations and difficult to
reach. The EPA has repeatedly stated that accessibility is a key factor in determining which
sources to require direct measurement. Of these sources, many of these sites are not easily
accessible to Yates staff.
Response: The commenter has misinterpreted the proposed rule, which does not require direct
measurement of leaks from sources listed under 98.233(q) for leak detection and leaker emission
factors. Instead, EPA is requiring leak detection at these sources using methods described in 40
CFR 98.234 (a). In todays final rule, EPA has not limited the leak detection methods to just
optical gas imaging devices. Instead, EPA has included techniques acceptable under 40 CFR part
60 (Method 21), and infra-red laser beam illuminated instruments; please see Section II.,
Petroleum and Natural Gas Systems, of the preamble for a response to this comment.
Commenter:
75 FR 18623: Use of Leak Detection and Leaking Component Emissions Factors
Comment: WBIH requests flexibility with the leak detection monitoring through the use of any
industry accepted instruments or practices. Optical gas imaging instruments are costly and
should not be mandated as the only monitoring method for leak detection when leak detection
with soap solution is an accepted, standard industry best practice. Also, EPA should not make an
assumption that a contractor would be hired when it is unknown if there is the required number
of contractors to provide leak detection with optical gas imaging instruments.
Response: EPA has included several additional leak detection techniques including Method 21
and infra-red laser beam illuminated instruments; please see Section II of the preamble for
further details. Regarding contractors to provide leak detection, EPA has revised the rule to
include Best Available Monitoring Methods. Please see Section II.F of the preamble to todays
final rule for further details.

1328
Duplicative provisions in proposed 98.233(q) should be deleted. Proposed 98.233 prescribes

various calculation methods for different emission source categories. Subparagraph (q) provides
certain calculation methods for sources listed in, inter alia, 98.232(f)(5), (g)(3), and (h)(4)
(generally relating to fugitive emissions). However, subparagraph (r) also provides calculation
methods for sources listed in 98.232(f)(5), (g)(3), and (h)(4). The duplicative coverage should
be eliminated, and the reference to sources listed in 98.232(f)(5), (g)(3), and (h)(4) should be
struck from subparagraph (q).
Response: EPA disagrees with the comment on duplication of emissions sources under
98.233(q) for leak detection and leaker emission factors and 98.233(r) for population count and
population emission factors. In 98.232(g)(3), (h)(4) and (f)(5) are listed equipment leak sources
that require either a leaker emission factor or a population emission factor.

Leaker versus population emission factors for underground storage: For natural gas underground
storage, fugitive emissions are identified as an emission source in 98.232 (f) (5). Based on
Table W-4, emission estimates employ population EFs for storage wellhead components and
leaker EFs for the balance of station components. Thus, 98.232 (f) (5) is cited in both 98.233
(q) for leak surveys and leaker EFs and 98.233 (r) for population EF-based estimates. To avoid
confusion on which EFs are used, which portion of the facility requires leak survey, and potential
double counting, additional clarity is desired. This could be addressed by adding the following to
the end of 98.233 (q) (4):
For underground storage facility components in storage wellhead service, population
emission factors apply under 98.233(r) and a leak survey is not required.
Response: EPA disagrees with the comment. Please refer to EPA-HQ-OAR-2009-0923-1061-5.

The requirements for leak detection and leaker emission factors would apply to emissions from
several source categories, including Sections 98.232 (f)(5),(g)(3), and (h)(4). These three sources
are also covered by Section 98.233(r) and should be deleted from this section to avoid double
reporting.
Response: EPA disagrees with this comment; please refer to EPA-HQ-OAR-2009-0923-1061-5.
1329

Leak detection and leaker emission factors
Section 98.233(q): The proposed rule requires that leak detection surveys be conducted annually
for fugitive emissions from specific source types within each sector. If a leak is detected, an
emission factor is applied as if the leak were occurring throughput the entire reporting year. In
order to encourage leak repairs by operators and to minimize GHG emissions, IPAMS requests
that EPA provide an optional alterative to annual leak detection surveys that would allow
reporters to conduct multiple surveys, as frequently as they deem appropriate. Accordingly, as
discussed on page 18623 of the preamble, emissions would reflect leak reductions as determined
by repairs and follow-up detection surveys.
To account for fugitive emissions reduction measures that the industry has undertaken in the last
few years since the leaker emission factors were developed, IPAMS requests that EPA provide a
provision in the final rule that allows leaker emission factors to be updated periodically, either by
EPA or by industry in consultation with EPA. As EPA stated on page 18622 of the preamble, the
proposed emissions factors will remain constant indefinitely until new factors are provided.
Response: EPA agrees with the commenter and will allow multiple leak surveys; please see the
response to EPA-HQ-OAR-2009-0923-1014-9. Concerning the application of updated emissions
factors, please see the response to EPA-HQ-OAR-2009-0923-1299-5.

Paragraph 98.233(q)(5) should be revised from LNG storage facilities . . . . connectors; and
other. to LNG storage facilities . . . connectors; and other fugitive sources.
Paragraph 98.233(q)(6) should be revised from "LNG imports and export facilities . . . .
connectors; and other." to "LNG imports and export facilities . . . connectors; and other fugitive
sources."
Response: EPA disagrees with the American Petroleum Institutes suggested amendments to the
rule text under Calculating GHG emissions Leak detection and leaker emission factors.
Other is a specific equipment type (i.e. equipment type other than connectors, pumps, or
valves) and is associated with an emission factor.
1330

Second, the proposal for leak detection and reporting would impose significant additional costs
and overstate the actual level of emissions. The proposed rule would require utilities to conduct
annual leak detection at above grade M&R stations thats metering and regulating and to
apply emission factors to those leaking sources. While current practices require regular leak
detection, the current level of detection does not nearly approach what would be required by this
proposal which calls for optical imaging. We question whether there would be enough equipment
or trained operators to conduct such annual surveys at all the M&R stations across the country
that would be affected.
Response: EPA disagrees with the comment regarding cost; please see response to comment
EPA-HQ-OAR-2009-0923-0049-7. In todays final rule, EPA does not require leak detection at
all meters and regulators at city gate stations; please refer to EPA-HQ-OAR-2009-0923-1065-4.
Please see the rulemaking docket (EPA-HQ-OAR-2009-0923), Understanding the Substance of
the DOT Regulations and Comparing Them to the Subpart W Requirements. EPA allows
additional leak detection techniques; please see Section II, Petroleum and Natural Gas Systems,
of the preamble for further details. Concerning the availability of equipment or trained
operators, EPA has allowed application for the use of best available monitoring methods; please
see Section II.F of the preamble to todays final rule.

EPA has not shown that LNG facilities emit significant levels of greenhouse gases from fugitive
sources. If a leak were to develop in an LNG facility in a cooldown process, the equipment
would be adjusted (e.g., tightening the flange) or the process would be stopped. If a safety relief
valve were to fail its annual test by failing to lift properly or failing to reseat, it would be
immediately replaced or repaired. If a water heater for a heat exchanger were to register the
presence of flammable gas in the water, indicating a small internal leak in a heat exchanger, the
heat exchanger would be identified and removed from service for testing and repair or
replacement. Leaks in LNG facilities are attended to promptly; leaker factors that presume that
leaks continue for 365 days give no recognition to the diligent operating and maintenance
practices required in, and common to, LNG facilities. Even if a facility were to count all possible
valves, pumps, connections, and other emission sources, and assume that all were detected
leaking, even applying the 365-day factors still would not yield a mass of emissions that
approached a reportable level. Indeed, facilities that have surveyed their plants as proposed have
identified emissions sources numbering in the single digits, an effort yielding a near-infinite cost
for the debatable benefit of identifying those tiny masses of emissions. It is likely EPAs
unfamiliarity with LNG facilities that has led to its proposed requirement of plant-wide fugitive
1331
detection surveys every year.

Response: EPA disagrees with the commenter on LNG facilities. LNG facilities handle natural
gas and hence are liable to equipment leaks just like in other natural gas handling facilities.
EPAs criteria for selecting segments and emission sources are outlined in the Greenhouse Gas
Emissions from the Petroleum and Natural Gas Industry: Background TSD (EPA-HQ-OAR
2009-0923-0027) under Selection of emissions sources for reporting. Data that is gathered
through the Mandatory Greenhouse Gas Reporting Rule Subpart W Petroleum and Natural Gas
Systems will be beneficial to inform future policy. EPA agrees that there is a lack of methane
emissions data from LNG facilities, which further necessitates collection of these data in order
for EPA to consider possible policy options. As a result, EPA has retained the requirements for
reporting from LNG facilities.
EPA agrees with the commenters point about improving the accuracy of emissions estimates to
account for leak repairs and is allowing multiple leak detection surveys to address this. Please
see the response to EPA-HQ-OAR-2009-0923-1014-9. In addition, EPA will evaluate the LNG
facility data received through the MRR, which are based on emission factors and methodologies
in subpart W. EPA will also evaluate other emission factors that EPA may receive. EPA may
also consider using direct measurement of equipment leaks and vents if EPA deems the LNG
facility emissions data received through the MRR is insufficient to inform policy.

Subpart W proposes that LNG facilities quantify certain specified process emissions, grouped as
defined into vented and fugitive categories. The facility is to measure emissions from certain
vented sources. Additionally, the EPA proposes that each facility identify 100% of its fugitive
emissions sources by examining all facility equipment with a prescribed detection instrument,
and prescribes leaker factors to estimate emissions from these identified sources. The EPA also
prescribes a population leaker factor for vapor recovery compressors; this type of plant system at
LNG facilities may take several forms, some of which are closed systems with no possibility of
leaks.
The EPA has chosen to apply these factors uniformly to generic equipment names across the
entire supply chain of petroleum and natural gas systems. However, given the extreme operating
conditions in LNG facilities, equipment is designed, selected, installed, operated, and maintained
to greatly reduce or eliminate leakage. The factors prescribed, which were introduced in 1993 for
the synthetic organic chemical manufacturing industry (SOCMI), have not specifically been
evaluated by the EPA for their applicability to current LNG facilities and equipment, nor for
their applicability for cryogenic service. Further, the EPA has overlooked the basic principle that
underlies the strong safety record of LNG facilities - to design, install, operate, and monitor the
facility to keep the product safely within the process equipment. This principle dovetails with
EPAs charge to minimize greenhouse gas emissions from that equipment in these facilities. (It is
also in LNG facilities commercial interest to minimize emissions, as the LNG, a premium fuel,
1332
as liquid or gas, is the facilities only product, and it must be dispatched from the plant to
customers in order to generate revenue.)
Response: EPA is aware of different technologies for vapor recovery compressors deployed by
industry across several sectors to reduce emissions, including the use of closed systems. If this
commenter is using low emission technologies as in a closed system, then EPA would expect to
see such emissions reflected in its submitted report. However, the use of low emission
technology is not known to be ubiquitous in the LNG industry; therefore, the collection of GHG
data from the LNG sector is important in understanding emissions from this sector to inform
future policy. As regards the prescription of SOCMI factors for applicability to LNG facilities
and equipment, in the absence of LNG-specific leak data, EPA sought a comparable factor that
would reduce burden on the industry while still capturing emissions information. Based on
expert judgment that they were the best publicly available factors, the SOCMI leaker factors for
components in NGL service were used to estimate LNG; however, the factors were adjusted for
the mass of methane. Please see the response to EPA-HQ-OAR-2009-0923-1299-11. Please also
see EPAs Greenhouse Gas Emissions from the Petroleum and Natural Gas Industry:
Background TSD (EPA-HQ-OAR-2009-0923-0027) for EPAs method for including LNG
facilities in the rule.

EPA Should Revise and Clarify Several Provisions for LNG Import Terminals (and for LNG
Storage Facilities If EPA Does Not Exclude Them)
EPA Should Not Apply Leaker Emission Factors at LNG Facilities Because Leaks Are Rare
and Fixed Immediately
With respect to proposed emission leaker factors, the EPA assumes that if a fugitive emission
source is detected, the source must have continued to leak for all 365 days of the year. This
hypothetical assumption would vastly overstate emissions and would not yield useful data for
policy development. Leaks are rare at LNG facilities. They are located and all but the smallest
are fixed as soon as they occur. As an example, a member recently performed a complete survey
of all of the equipment, and piping systems at their LNG storage facility using the FLIR camera.
The member only found one small leak at a rod packing on the refrigeration compressor where
mechanics were working, as was expected. These facilities are very tight by design and
operation. LNG facilities cannot and do not permit conditions such as LNG valve leakage to
continue after detection. Detection of leaks is rapid due to installed gas detection and physical
facility inspections by qualified operating staff. When leaks are detected, they are generally
corrected immediately. Thus, using a leaker emission factor that assumes the leak continues all
year would seriously overstate fugitive emissions at an LNG storage facility or import terminal.
1333
Response: EPA disagrees with the comment to not quantify LNG facility equipment leaks.
Please see the response to EPA-HQ-OAR-2009-0923-1299-11. EPA will allow multiple leak
detection surveys; please see the response to EPA-HQ-OAR-2009-0923-1014-9.

LNG Valves and Minimal Valve Leakage.
Due to the combined conditions of extreme temperature and pressure, valves (gate, ball,
butterfly, etc) in LNG service are typically designed with extended bonnets and utilize multiple
rings of V-ring style stem packing, made typically of PTFE (Teflon). This style of packing is
very resilient and has high sealing qualities, thereby greatly minimizing if not eliminating
fugitive emissions. Furthermore, since valves in LNG or natural gas service can range from
to 24 or greater diameter; the use of a required single leaker factor appears arbitrary.
Response: EPA has reviewed this comment and disagrees. EPAs decision to include LNG
storage facilities was based on several considerations. First, please see EPAs Greenhouse Gas
Emissions from the Petroleum and Natural Gas Industry: Background TSD (EPA-HQ-OAR
2009-0923-0027) for EPAs method for including LNG storage in the rule. In addition, EPA is
aware of different technologies for valves deployed by industry across several sectors to reduce
emissions. If this commenter is using low emission technologies, then EPA would expect to see
such emissions reflected in their submitted report. However, the use of low emission technology
does not eliminate the need for LNG storage facilities to report; as such information may
demonstrate differentiated emissions levels, which would inform future policy. EPA disagrees
with the comment stating that the choice of a leaker factor was arbitrary, as EPA used the best
publicly available emission factors.

Organization: Sempra LNG
Commenter: David M. Cobb
Table W-6 identifies a single emissions factor for valves. Due to the combined conditions of
extreme temperature and pressure, valves (gate, ball, butterfly, etc) in LNG service are typically
designed with extended bonnets and utilize multiple rings of V-ring style stem packing, made
typically of PTFE (Teflon). This style of packing is very resilient and has high sealing qualities,
thereby greatly minimizing if not eliminating fugitive emissions. Furthermore, valves in LNG or
natural gas service can range from 1/4" to 24" or greater diameter. The use of a single factor
appears arbitrary and may result in a significant overstatement of fugitive emissions for LNG
facilities.
1334
Response: EPA disagrees with the comment on the use of a single emissions factor. Please see
the response to EPA-HQ-OAR-2009-0923-1016-42.

LNG Valves and Minimal Valve Leakage
Due to the combined conditions of temperature and pressure, valves (gate, ball, butterfly, etc.) in
LNG service are typically designed with extended bonnets and utilize multiple rings of V-ring
style stem packing, made typically of PTFE (Teflon). This style of packing is very resilient and
has high sealing qualities, thereby greatly minimizing if not eliminating fugitive emissions.
Furthermore, since valves in LNG or natural gas service can range from 1/4 to 24 or greater
diameter; the use of a required single leaker factor appears arbitrary.
Response: EPA disagrees with the comment on the use of emission factors. Please see the

LNG Safety Relief Devices
Due to extreme operating conditions, the majority of safety relief valves in LNG service are soft
seated style valves, again this greatly minimizes, if not eliminates, leakage at pressures below the
valve opening set point. Additionally, relief valves in the LNG industry in the U.S. are
maintained at a very high level of performance, as each relief valve is required by the federal
LNG code, 49 CFR 193.2619, to be inspected and tested annually for lift pressure and positive
reseating, which indicates that it operates properly and is not leaking. PHMSA inspects safety
valve testing records to confirm that the facilitys annual testing of valves has met the regulatory
requirements for test frequency and valve performance.
Response: Please see the response to EPA-HQ-OAR-2009-0923-1016-42. EPA is not excluding
LNG facilities from reporting in todays final rule because they are regulated by the PHMSA.
Please see the response to EPA-HQ-OAR-2009-0923-1299-11 and see the rulemaking docket
(EPA-HQ-OAR-2009-0923), Understanding the Substance of the DOT Regulations and
Comparing Them to the Subpart W Requirements.

1335

LNG Safety Relief Devices.
Due to extreme operating conditions, the majority of safety relief valves in LNG service are soft
seated style valves, again to greatly minimize if not eliminate premature leakage at pressures
below the valve opening set point. Additionally, relief valves in the LNG industry in the U.S. are
maintained at a very high level of performance, as each relief valve at LNG facilities is required
by the federal LNG code, 49 C.F.R 193.2619, to be inspected and tested annually for lift
pressure and positive reseating, which indicates that it operates properly and is not leaking.
PHMSA inspects safety valve testing records to confirm that the facilitys annual testing of
valves has met the regulatory requirements for test frequency and valve performance.
Response: Please see the response to EPA-HQ-OAR-2009-0923-1016-42. EPA is not excluding
LNG facilities from reporting in todays final rule; please see the response to EPA-HQ-OAR
2009-0923-1299-11. Please see the rulemaking docket (EPA-HQ-OAR-2009-0923),
Understanding the Substance of the DOT Regulations and Comparing Them to the Subpart W
Requirements.
13.1.3 ALTERNATIVE OPTIONS TO CONSIDER

Organization:
Allowing choice of Emissions Factors or Direct Quantification (98.233 (q) and 98.233 (r))
EPA previously recognized the importance of allowing the use of either emission factors or
Direct Quantification of fugitive emissions.
EPA Document EPA-453/R-95-017, 1995 Protocol for Equipment Leak Emission Estimates,
presented four methods for reporting fugitive hydrocarbon emissions. The first three methods
used generalized emissions factors based on population counts or leaker counts. The fourth
method allowed a facility to quantify leaks to obtain better accuracy in emissions reporting. If
you were to allow facilities to quantify leaks, facilities with large leaks would report higher
emissions than predicted by Tables W-1 through W-7. Facilities with small leaks would report
lower emissions than predicted by Tables W-1 through W-7.
This would have the double advantage of (1) reporting measured emissions for each facility
(rather than calculated emissions), and (2) motivating the facilities to fix large leaks, thereby
lowering emissions to the environment.
It would also allow companies to become industry-leaders through proactive control of
fugitive emissions.
1336
A review of the data in my document, which you referenced, will show how and why Direct
Quantification will decrease the burden on the environment.
Pages C-1 through C-4 of the report list the individual leak rates of 176 valves in gas service at
Gas Processing Plants. The individual leaks are shown in the table below.
Gas Valve Leak Rates in Emission Factors for Oil and Gas
Production Operations, January 1995 (scf/hr)
As can be seen, most of the leaks are small (less than 1.00 scf/hr) and a few are large (greater
than 10.00 scf/hr). This population pattern is typical of all leak data sets, regardless of facility
type, component type, stream composition or pressure.
Total emissions from the 176 valves are 542 scf/hr. The largest leak (121.99 scf/hr) is 23% of
that total. If a facility were to repair that single valve, the burden on the environment would be
reduced by 23%. Repairing the 10 largest leaks would reduce the burden by 62%.
Using Direct Quantification to prioritize maintenance work is essential to controlling fugitive
emissions and documenting their reduction.
For this reason, I recommend allowing facilities to make Direct Quantification of the leaks if
1337
they so choose.
The proposed Subpart W already mandates Direct Quantification for Reciprocating Compressor
Rod Packing Venting (98.233 (p) (3) (ii)); I am suggesting that Direct Quantification be
included as an option for all facilities in 98.230.
Response: EPA disagrees that it should provide an option for reporters to submit measured data
for equipment leak emissions. In the April 2009 proposed rule, EPA required reporters to
measure their equipment leaks; however, many potential reporters commented that the cost
burden was too high. EPA accepted these comments and revised the rule to include population
emission factors and leaker factors to reduce compliance burden on industry while maintaining
the necessary quality of data to inform policy. Please see Section III of the preamble to the April
2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002) and response to comment
EPA-HQ-OAR-2009-0923-1155-12.

Organization:
In 98.233 (q) and 98.233 (r), allow affected facilities the choice of either calculating fugitive
emissions using Emissions Factors or measuring fugitive emissions using Direct Quantification
Response: EPA disagrees that it should provide an option for direct measurement as an
alternative to monitoring equipment leaks; please see the response to EPA-HQ-OAR-2009-0923
0132-3.

Organization: ENVIROTECH ENGINEERING
Commenter: Terence Trefiak
Using general emission factors can cause significant error in estimating emission rates.
In the example above these two leaking components would be given the same estimated leak rate
which could be inaccurate (over estimation or under estimation) by many orders of magnitude.
In our experience the cost of leak measurement is approx. 10% of the total cost of performing a
fugitive emission assessment. The majority of the assessment time is used to inspect and detect
emissions and the time required to measure leaks is relatively low in comparison, but the value
of having the actual leak rate is significant. All of our clients to date have requested leak rate
direct measurement as part of the fugitive emission assessments.
We recommend that, if a reporting company has undertaken leak rate measurements as part of a
fugitive emission assessment, that the finalized GHG Reporting Rule allows the option for the
reporter to use the measured information as opposed to having to use the prescribed generic
leakage factors.
1338
Response: EPA disagrees that it should provide an option for direct measurement as an
alternative to monitoring equipment leaks. Please see the response to EPA-HQ-OAR-2009
0923-0132-3.
WCI Recommendations: WCI recommends adoption of USEPA methods 98.233(q) and (r) with
minor modifications. While there are methods available that provide quantitative data on fugitive
leaks (e.g. use of calibrated sampling bags and/or high-volume sampler) these methods are costly
and time consuming. Further more, these quantitative methods provide a single snapshot in time
and no information concerning leak duration.
The draft EPA rule quantification methodologies will not produce cap-and-trade quality data
because they use default emissions factors rather than actual measurements of emissions.
Additionally, these calculation methods assume that the component has been leaking for the
entire reporting period and this assumption will overestimate actual emissions.
While the data generated are not cap-and-trade quality, section 98.233(q) does require that
reporters conduct an annual component screening (a directed inspection program) on a subset of
components. This will provide operators with operable information which both identifies leaking
components and provides semi-quantitative data on leak rates. This information will help
operators identify and prioritize leak mitigation opportunities.
WCI Modifications: Equations W-18 and W-19 contain errors which should be corrected. In both
equations the variable GHGi is set equal to 1 for many sources. GHGi represents the
concentration of GHGi (where i = CH4 and CO2) in the gas stream in question. Setting GHGi
equal to 1 for CO2 will result in a large overestimation of CO2 emissions. For CO2, GHGi should
be the concentration of CO2 in the gas stream in question.
Response: Given that the purpose of this rule is to collect data to inform a variety of programs
and policies, EPA has determined that the methodological rigor presented in todays final rule is
sufficient for these purposes. If and when a particular policy is implemented (e.g., cap and
trade), EPA will evaluate whether adjustments to the methodologies in this subpart are
warranted. Please see the response to EPA-HQ-OAR-2009-0923-0582-10.
EPA agrees that the rule requirements should provide the option to account for the duration that a
component leaks rather than assuming that it leaks for the entire reporting period. However,
EPA is not requiring that reporters perform multiple leak surveys each year. Todays final rule
allows reporters the option to account for the time a component leaks and provides guidance on
how to estimate this based on the number of leak screening surveys that the reporter chooses to
conduct.
EPA agrees that in the proposed rule the default fraction of CO2 for equipment leaks was
inaccurate. Todays final rule has been revised and has a more accurate parameter for the CO2
1339
fraction of emissions using the same method as the U.S. Inventory of Greenhouse Gas
Emissions and Sinks: 1990 2008. For onshore petroleum and natural gas production and
onshore natural gas processing, reporters use their own CO2 concentration from composition
analysis. For pipeline quality natural gas in transmission, underground storage, LNG storage,
LNG import and export, and distribution, the CO2 content is 1.1x10-2.
Section 98.233(a) Natural gas pneumatic high bleed device venting
Sections 98.233 (a) and (b) require the use of either manufacturers data for high bleed
pneumatics or the application of an emission factor for low bleed pneumatics. However,
identifying the model of a particular device may be difficult if name plates are obscured by paint
or wear. Additionally, some units may have been modified over time to reduce their bleed rate,
but accurate records of these modifications may not be available.
The EPA should provide a default leak rate that can be used for those devices where the
manufacturer or model number is not known. Alternatively, the EPA could allow operators to
use appropriate factors from sources such as the API Compendium to obtain factors for these
particular devices. As an additional option, the Hi-Flow sampler or similar device could be used
to make a measurement of the bleed rate from pneumatic devices. If companies already have this
technology in house or are having it contracted for other measurements at their sites, this could
provide a method to obtain data that might otherwise be hard to get. The key here is to ensure
flexibility in monitoring options.
Response: EPA agrees with the comment regarding the difficulty of identifying pneumatic
device manufacturers. Todays final rule has been revised to use a population emission factor for
high and low-bleed pneumatic devices, and therefore identifying manufacturers and
manufacturer data is no longer necessary. Please see Section II.F of the preamble to todays final
rule.
98.233 (a): CAPP would like clarity on the following statement in sub-section (2): Who decides
what is a similar device model? Is this a role the EPA would take on an on-going basis and
how would a facility go about receiving a decision on a similar device model?
To reduce the burden associated with collecting data on the number of hours each controller was
in operation in the year, CAPP suggests that as a default, companies assume that all controllers in
1340
this category are in operation 24 hours per day, 365 days per year. This would provide a
conservative estimate of emissions from this source. Those who have or choose to collect actual
operating hours would be free to use those data in lieu of the default. This method would then be
consistent with 98.233(b).
Where manufacturer specific bleed rate in not available, CAPP also recommends allowing the
use of average gas consumption rates for high bleed pneumatic instruments (e.g., CAPP 1993
Table 2-2) and that the EPA may want to consider using generic count based on facility and
process equipment type as per CAPP 2005, Volume 5.
Response: EPA agrees and todays final rule has been revised to use a population emission
factor for low and high-bleed pneumatic devices. Please see the response to EPA-HQ-OAR
2009-0923-1011-37. Todays final rule has also been revised such that pneumatic devices will
report with an assumption of 8,760 operating hours per year. EPA determined that monitoring
the operational time for each pneumatic device is overly burdensome; please see Section II.F of
CAPP supports the EPAs decision to provide alternatives to the use of direct measurement, such
as engineering estimates, in recognition of the fact that the installation of new meters or emission
monitoring equipment often requires a facility shutdown in the oil and gas sector (as in many
other industrial sectors). These shutdowns typically only occur on an annual or semi annual
basis, making it difficult to ensure compliance without costly unscheduled shutdowns.
Response: EPA agrees and has maintained these provisions in todays final rule.

Leak Detection Surveys and Annual Calculations
EPA is requesting comments on an alternative approach for fugitive leak detection and reporting.
This approach includes requiring a facility to conduct multiple surveys and would take into
account any repairs that were conducted between surveys. Red Cedar recommends that EPA
allow this alternative approach as an option in the final rule, but does not recommend that it be
mandatory. For a facility that makes repairs to reduce fugitive emissions this option would allow
credit to be given for those repairs. By making it an optional approach, facilities that do not make
repairs, likely due to few leaking components, would not be burdened with multiple surveys that
would likely show very similar results.
1341
Response: EPA agrees and has revised todays final rule to allow reporters the option to
perform multiple leak surveys and repair discovered equipment leaks, and account for those
repairs in their reported emissions. Please see the response to EPA-HQ-OAR-2009-0923-1015
39.
Organization: GDF SUEZ Gas NA LLC
Commenter: Francis J. Katulak
In summary, the extensive requirements of existing federal LNG regulations, the robust
equipment designed and selected for LNG facilities, the principles and practices developed by
the LNG industry and by individual LNG facilities, and the regular inspections and evaluations
by federal inspectors all support the efforts of facility personnel to ensure that LNG facilities are
operated and maintained to keep the LNG safely within the process equipment. As a result,
fugitive emissions of greenhouse gases are likewise nearly eliminated. Therefore, EPAs
proposal to require a burdensome optical detection survey to attempt to detect fugitive leaks in
an LNG facility cannot be justified, due to the high cost and the insignificant findings that would
result from using EPAs methodology. And it is inappropriate to compute emissions based on
assumed 365-day leaks, to apply factors that relate to equipment (and possibly fluids) distinct
from those in LNG facilities, and to use an instrument for detection that has not been shown to
provide accurate identification of emissions sources (and which is so specific as to be sure to be
superseded in the future, necessitating a change to the regulation), rather than allowing those
facilities where a fugitive emissions survey is appropriate to select from multiple approved
detection methods to identify active emission sources.
Response: EPA disagrees, and will continue to retain reporting requirements from LNG
facilities. Please see response to comment EPA-HQ-OAR-2009-0923-1025-1.
EPA disagrees that optical gas imaging instruments will be unable to detect methane leaks from
LNG equipment and that the fluid properties governing the ability of a substance to leak through
an orifice will differ substantially between LNG and NGLs such that the emission factors cannot
be used to scope emissions with the intention to inform future policy. In the absence of LNGspecific leak data, EPA sought a comparable factor that would reduce burden on the industry
while still capturing emissions information. Based on expert judgment, the SOCMI leaker
factors for components in NGL service were used to estimate LNG; however, the factors were
adjusted for the mass of methane.
EPA also must collect data from the LNG sector in order to inform possible future policy. The
information on emissions from the LNG sector are necessary for EPA to evaluate the source and
its level of emissions to determine what, if any, action is appropriate to take regarding those
emissions. Furthermore, as explained in Section II.H of the preamble for the 2009 final
Mandatory Reporting of Greenhouse Gases and 40 CFR 98.2(i), if a reporters emissions are
below 15,000 mt CO2e for three years in a row or below 25,000 mt CO2e for five years in a row
they can cease reporting.
1342

(Preamble pp. 79-80) The API compendium emissions factors that we are proposing to use in the
upstream oil and gas production sector may be underestimating emissions. EPA seeks comment
on how to improve these factors and/or collect more accurate data.
API supports the use of emission factors from the API Compendium, as these are widely used
and accepted by the oil and natural gas industry worldwide. In addition, Subpart W should
provide for the use of new emission factors as they develop. As such, API recommends
removing Tables W-1 through W-7 from the reporting rule and instead reference a technical
document of emission factors that is more readily updated.
Section VIII of this document details a comparison between the emission factors proposed for
Subpart W and the API Compendium. Errors and recommendations for specific emission sources
are noted there. For many source types, the API Compendium emission factors are larger or
equivalent to those proposed by EPA.
Response: EPA disagrees with only using emission factors from the API Compendium; please
refer to EPA-HQ-OAR-2009-0923-3524-4 for further details. EPA will not remove the emission
factors from subpart W and place them in another reference technical document, as this will not
reduce the amount of time needed to reference updated factors as compared to having the factors
remain in the rule itself.
Organization: Spectra Energy Corp
Commenter: Brianne Metzger-Doran
Vent Measurement for Manifold Vent Lines on a Continuous Basis Should be Allowed as an
Operator Option
Continuous measurement of manifolded vent lines and aggregate annual emissions reporting
should be allowed, at the operators discretion, as an approved alternative to the methods defined
in 98.233. This continuous aggregate reporting option, supported by INGAA, is of particular
importance to Spectra Energy because Spectra Energy hasat substantial cost and at a
substantial number of its facilitiesinstalled manifolded vent lines capable of supporting
reliable continuous emissions measurement devices. These manifolded vent lines, or common
headers, were installed to satisfy the source control requirements of the 1989 PCB consent
decree between the United States and Texas Eastern Transmission Corporation. These common
headers provide substantial environmental benefit and allow Spectra Energy to continue its PCB
source control activities by directing the vented streams to liquid separators prior to venting to
the atmosphere. As currently contemplated by 98.233, the Proposed Rule would require the
measurement of each covered emissions source connected to the common header, which could in
1343
some cases require Spectra Energy to physically dismantle the manifolded vent line system at
least once every year to safely obtain snapshot emission measurements. If continuous
measurement of manifolded vent lines and aggregate annual emissions reporting is allowed as an
operator option, Spectra Energy, and operators like Spectra Energy, will be able to safely collect
and report to EPA continuously measured reliable data. This data set will also allow EPA to
compare these results with the GHG emissions data reported by operators that take a snapshot
measurement of emissions from individual vents or pieces of equipment.
Response: EPA disagrees with the comment that compressor venting should be allowed to
conduct direct measurement of common manifolded vent lines. The rule allows installation of a
port on each vent line to the manifold, for insertion of a temporary meter, and these ports can be
installed at ground level. Typically, compressor venting emissions vary with the mode of
operation of the compressor. The emissions are highest when the compressor is operating and
lower when they are in the not operating depressurized mode. However, when the compressor is
not operating depressurized, there may be leakage of natural gas through the unit isolation valve,
particularly if the valve seat has become fouled and will not completely close. Hence, to
correctly characterize annual emissions from compressors, estimation of emissions in different
compressor modes is required. A temporary meter such as a vane anemometer or permanent
meter such as an orifice meter can be used to measure emissions from the vents.

Further, it is strongly recommended that EPA approve alternate methods of determining and
reporting fugitive emissions on reciprocating compressor rod packings other than direct
measurement, such as manufacturer specifications, engineering calculations, or industry
benchmark values for comparable engines. This will remove the financial burden associated with
conducting leak testing on every reciprocating compressor rod packing nationwide. For one
facility with six compressors having a total of eighteen compression cylinder rod packings, we
have received an estimate of $40,000 to $60,000 per year to conduct the prescribed annual
sampling. Obviously, the testing can be done, but the result of this sampling is of little value
because it only captures fugitive emissions, if any, at one point in time, since any leaking rod
packings are promptly addressed as part of the routine maintenance of these units.
Response: EPA has determined that commenters incorrect assumptions and misinterpretations
of the proposed rule largely account for the discrepancy in compressor monitoring cost burden
estimated by EPA. The rule does not require that vent manifolds be broken apart. The rule
allows installation of a port on each vent line to the manifold, for insertion of a temporary meter,
and these ports can be installed at ground level. For more information, please see the response to
EPA-HQ-OAR-2009-0923-1099-18. Further, the sampling is not of little value. EPA disagrees
with the commenter that leaking rod packings are promptly addressed. On the contrary, EPA
has learned that rod packing emissions varies greatly depending on the compressor itself and the
wear that a specific compressor may have on rod packing. Also, compressor maintenance is
1344
most often done at fixed time intervals, not based on the wear of the rod-packing which impacts
emissions rates and creates substantial inconsistencies in emissions from this source.

Second, while GPA supports the full use of the API Compendium to calculate GHG emissions
for all sources covered by the proposed Subpart W, at a minimum we suggest the application of
the API Compendium in place of any direct measurement methods. For the gas gathering and
processing sector, direct measurement is generally proposed to be limited to compressor wet seal
vents and rod packing venting. The current proposal potentially requires direct measurement of
compressor rod packing vents from every cylinder of a compressor in cases where the vent lines
are not tied to a common line, which is a common practice in gathering compression facilities
and production facilities. The prescribed methods for these sources are flawed because they
assume a snapshot measurement applied across an entire year is more accurate than an emission
factor based on well documented and controlled studies.
Response: EPA disagrees that the API Compendium is suitable for all sources covered by
subpart W for the intent of informing future policy. Please see the response to EPA-HQ-OAR
2009-0923-3524-4.
With regard to monitoring of gathering and boosting emissions, todays final rule no longer
requires reporting of those sources; please see Section II.F of the preamble to todays final rule
for the response to this comment.
EPA disagrees that direct measurement of emissions from compressor venting is less accurate
than emission factors. With regard to centrifugal compressor wet seals, please see the response
to EPA-HQ-OAR-2009-0923-1206-45. With regard to reciprocating compressor rod packing,
commenters made incorrect assumptions about measuring manifolded vents. The rule does not
require that vent manifolds be broken apart. The rule allows installation of a port on each vent
line to the manifold, for insertion of a temporary meter, and these ports can be installed at ground
level. For more information, please see the response to EPA-HQ-OAR-2009-0923-1099-18.

Organization:
Commenter: Matt Harrison
Please allow the use of emission factors for compressor related components. EPA should not
require direct measurements for all gas compressors. As the project manager of the GRI/EPA
studies from the 1990's that are cited in the proposed rule, I am familiar with many of these
emission measurements. The direct compressor measurements EPA is requiring for five fugitive
1345
and vented components at compressor stations are more technical, complex, and expensive to
measure than EPA has estimated. Some emission sources are not possible to individually
measure without physical modifications to the site. EPA should continue to fund research that
refines the key natural gas emission factors. EPA should not require the entire industry to make
measurements for a GHG reporting rule. EPA could allow for direct measurement as an
alternative to emission factors, if the reporter elects, in their sole discretion.
Please allow for the use of scaled emission factors, rather than more detailed modeling that the
proposed rule requires for sources like glycol dehydrators, amine units, and tanks. For the
reporting rule purposes, the accuracy of scaled emission factors should be acceptable.
Response: EPA disagrees that direct measurement of emissions from compressor venting should
be replaced by emission factors. With regard to centrifugal compressor wet seals, please see the
response to EPA-HQ-OAR-2009-0923-1206-45. With regard to reciprocating compressor rod
packing, please see the response to EPA-HQ-OAR-2009-0923-1099-18. The rule does not
require that vent manifolds be broken apart. The rule allows installation of a port on each vent
line to the manifold, for insertion of a temporary meter, and these ports can be installed at ground
level.
EPA disagrees that scaled emission factors should be used for glycol dehydrator, amine units,
and storage tank emissions. With regard to storage tanks, EPA designated it as a source with
large uncertainty using typical emission factors, as discussed in Greenhouse Gas Emissions
Reporting from the Petroleum and Natural Gas Industry Background Technical Support
Document (EPA-HQ-OAR-2009-0923). With regard to acid gas removal vent, please see
response to comment EPA-HQ-OAR-2009-0923-1024-26. With regard to glycol dehydrators,
EPA determined that reporters will obtain better emissions estimates from larger glycol
dehydrators using simulations versus emission factors, and without a substantial increase in cost
burden. Once the initial set up for the software is accomplished, it takes little time to make
multiple runs with new inputs. However, for small dehydrators less than 0.4 million cubic feet
per day throughput, and storage tanks with separator or wellhead production throughputs less
than 10 barrels per day, EPA agrees that simulations are an unnecessary burden for the
magnitude of emissions, and has revised the rule to provide default emission factors that may be
used. Please see Equipment Threshold for Dehydrators and Equipment Threshold for Tanks
in the rulemaking docket (EPA-HQ-OAR-2009-0923) and Section II.E of the preamble to
todays final rule.
For small emission sources, the use of industry standard emission factors. For example:
For glycol dehydrators with throughput less than 3 MMcf/day, appropriate (i.e. considering use
of flash separator and gas-assisted glycol pumps) emission factors from the API Compendium
would be used rather than collect all the data and samples required for a GLYCalc estimation.
While the emission factor approach may not be accurate for some individual sources, for the
1346
population of sources errors average out and the emission factor estimates would provide quality
data to inform policy decisions.
Response: EPA agrees that small glycol dehydrators may use emission factors to estimate
emissions. Please see Equipment Threshold for Dehydrators in the rulemaking docket (EPA
HQ-OAR-2009-0923).
Less Direct Measurement Should be Required.
Although EPA has reduced some direct measurement requirements from those included in its
April 2009 MRR Subpart W proposal, EPA's current proposal does not acknowledge
longstanding and accepted international protocols adopted by other countries to calculate
emissions from the oil and gas sector. These protocols include Australia's National Greenhouse
and Energy Reporting System ("NGERS"), the European Union Emission Trading Scheme ("EU
ETS"), and Alberta's Greenhouse Gas Reporting Program, which all either allow or require
reporting emissions calculated using the Compendium. 191 EPA's proposal assumes that
calculating and reporting emissions from each small component related to oil and gas production
will result in a more accurate GHG inventory, rather than using carefully documented emissions
factors applied across entire facilities. However, utilizing burdensome methods of direct
measurement will not necessarily result in an emissions inventory of greater value or validity, as
compared to using other accepted and proven techniques, such as the Compendium. In fact, the
CO2 emissions estimated from onshore facilities using the Compendium have been shown to be
comparable to other available methods. 192 Moreover, if all of industry were to use the
Compendium, the uncertainty in the values reported would be more uniform across the reporting
industry, whereas if direct measurement were used, the uncertainties would be non-uniform
across the reporting industry. Uncertainties can arise when using models or equations, but these
errors can be eliminated in advance through the proper application of methods to the source
191
Australian Government, Department of Climate Change, Technical Guidelines for the estimation of greenhouse
gas emissions by facilities in Australia, ch. 3 (June 2009); Answers to Frequently Asked Questions on Greenhouse
Gas Emissions Monitoring and Reporting under the EU Emissions Trading System Pursuant Directive
2003/87/EC, at 13 (Sept. 2007); Government of Alberta, Technical Guidance for Completing Specified Gas
Baseline Emission Intensity Applications, at 30 (Feb. 2010).
192
NARSTO Emission Inventory Workshop, Development of the API Compendium for Estimating Greenhouse Gas
Emissions (Oct . 2003) , available at _http://nas
.carer.uiowa.edu/ICARTT/Seminars%20and%20Formal%20Presentations/NARS"TO Emissions workshop
2003/presentations/ritter .pdf (comparing the Compendium to the following other methods: Exploration and
Production Forum, Methods for Estimating Atmospheric Emissions from E&P Operations (E&P Forum 1994);
Intergovernmental Panel on Climate Change, Guidelines for National Greenhouse Gas Inventories (IPCC 1997;
UNECE/EMEP 1999; IPCC 2001); Regional Association of Oil and Natural Gas Companies in Latin America and
the Caribbean, Atmospheric Emissions Inventories Methodologies in the Petroleum Industry (ARPEL 1998);
World Resources Institute and World Business Council for Sustainable Development, The Greenhouse Gas
Protocol ffRI/WBCSD 2001)).
1347
category. LEVON Group, Addressing Uncertainty in Oil and Natural Gas Industry Greenhouse
Gas Uncertainties, at 2-1 to 2-2 (Sept. 2009). 193 In contrast, "the nature of the measurement
process. . . makes it impossible to measure a physical quantity without error." Id. at 2-2. Errors
may originate from "sampling, measuring, incomplete reference data, or inconclusive expert
judgment." Id. at 2-2. Thus, "determining the true value of any measured variable is not practical
due to the limitations of measurement equipment and procedures, and the possibility of human
error." Id at 3-1 . Therefore, by having sources use the Compendium, the uncertainty of the
results reported should be uniform and the accuracy and overall reliability of the data should
increase, at a fraction of the cost.
Response: EPA disagrees that applying API Compendium emission factors across the industry
would result in more accurate emissions estimates than the methods in todays final rule. The
emission factors in the US Inventory, and also in the API Compendium, have been demonstrated
to underestimate emissions by an order of magnitude or more for sources including gas well
liquids unloading, gas well completions and workovers, and centrifugal compressor wet seal
degassing vents. EPA discusses the methodologies it considered and its goals in the Greenhouse
Gas Emissions Reporting from the Petroleum and Natural Gas Industry Background Technical
Support Document (EPA-HQ-OAR-2009-0923).
As explained in Section 6, above, the rule should allow for the use of API Compendium factors
to estimate emissions from centrifugal compressor wet seal degassing vents.
Response: EPA has reviewed this comment and disagrees with this comment. Please see

As we stated in our 2009 comments, there are clearly emissions estimating tools available that
have been used and can be improved without imposing these new requirements. If better
estimates are needed for this small portion of the GHG inventory, EPA can turn to its own
capabilities to address these estimates. EPA operates the Natural Gas Star program. It cites
information in supporting documents to the 2009 proposal indicating that Natural Gas Star has
identified and to some degree determined what emissions areas at production systems generate
the most emissions. Similarly, API released a new version of its Compendium of Greenhouse
Gas Emissions Estimation Methodologies for the Oil and Gas Industry. These tools can be used
193
Prepared for the International Petroleum Industry Environmental Conservation Association ("IPIECA"), API, and
Concawe.
1348
to create reasonable average emissions projections for production systems that could be linked to
production volumes. And, EPA could then improve its GHG estimates for onshore petroleum
and natural gas production without imposing the costly reporting burdens that would result from
inclusion of these operations in the reporting requirements.
Response: EPA has reviewed this comment and disagrees with applying API Compendium
emission factors across the industry. Please see response to comment EPA-HQ-OAR-2009
0923-3524-4. The API Compendium methods underestimate emissions substantially for sources
such as gas well liquids unloading, gas well completions and workovers, and centrifugal
compressor wet seal degassing vents. Therefore, it is necessary to gather actual emissions data
from these and other sources such that EPA can meet data quality standards necessary to inform
policy.
Provide alternative to optical gas imaging instrument - Limiting leak identification to only an
optical gas imaging instrument is too restrictive and would result in increased cost and will
worsen the shortage of qualified personnel to conduct leak detection. Options should be offered
to use alternative methodologies that are available, currently are in practice, and provide better or
equivalent results in most cases.
Response: EPA agrees that alternatives to optical gas imaging may be used for leak detection
for accessible components. Please see the response in Section II.F of the preamble to todays
final rule.
In addition to removal of unnecessary emission sources, Nobles recommends that the Proposed
Rule revisions to reduce burden while collecting GHG emissions data to develop a representative
inventory include alternative, simpler, streamlined GHG emission estimation methods:
Representative sampling and measurements to develop emission factors / data to estimate entire
population emissions rather than test or sample every emission source
Response: EPA compared alternative monitoring methodologies to determine which sources
require direct measurement or leak detection and if a representative sample would result in
reduced burden on industry while maintaining the necessary quality of data to inform policy.
EPA determined that direct measurement of representative samples was appropriate for gas well
venting during well completions or workovers with hydraulic fracturing, centrifugal compressors
wet seal degassing venting, and reciprocating compressor rod packing venting. EPA also
1349
determined that leak detection of representative samples was appropriate for above-ground
meters and regulators at non-custody transfer city gate stations.

98.233 Calculating GHGs emissions.
ConocoPhillips General Comment:
We request EPA add an option that applies to all sources that gives the reporter the ability to
submit to EPA a request for a waiver from the prescribed calculation methodology for a
particular source. The waiver request can require the reporter to justify an alternative calculating
methodology. This will allow for circumstances where an alternative method is more accurate for
GHG emission calculations while also reducing the data collection/reporting burden. In addition,
the approved methodology will be documented in the GHG Monitoring Plan.
Response: EPA disagrees that reporters should have the ability to request a waiver from the
required calculation methodology for a particular source. EPA selected the monitoring
methodologies that would minimize burden on industry while maintaining the necessary quality
and consistency of data to inform policy. Additionally, we have provided flexibility by
introducing simplified methods and best available monitoring methods. Please refer to Section
II.E and II.F of the preamble to todays final rule. Based on the changes made, we have
concluded that it is not reasonable to review and regulate customized, individual requests for
each source from each reporter. Furthermore, to ensure data consistency and quality, it is
necessary to have all reporters following the same methodologies for the same emission sources.
13.2
GREENHOUSE GAS CALCULATION PROCEDURES
Commenter:

Screening Criteria and Levels are Necessary to Establish the Requirement to Report under
Subpart W.
Based on the currently proposed calculation methods in Subpart W, emissions from many of the
GHG sources are unpredictable, which will make determining applicability relative to the
25kTon/yr threshold impractical. In particular, emissions from the following sources cannot be
predicted: blowdowns, flares, compressor vents & seal, and fugitive leaks. Due to the
1350
unpredictable nature of these sources, a facility will not be able to establish applicability to
Subpart W reporting until well into a calendar year and perhaps not until the end of a calendar .
Also, for compressor vents & seals and fugitive leaks, a facility will not be able to determine
emissions from the sources until the direct detection and measurement is completed. In order for
the regulated community to establish applicability of the GHG reporting rule, specifically for
Subpart W, screening criteria need to be established that allows this determination to be made
with certainty in January of each calendar year.
Response: To assist reporters in determining applicability, EPA plans to provide reporters with
a screening tool. Please see Section II.F of the preamble to todays final rule for a response to
this comment and further details on the screening tool.
Allow leak measurements as an option to determine bleed rates of pneumatic control devices.
Response: EPA has determined that leak measurement to determine the bleed rates of
pneumatic control device is too onerous and has chosen not to include it in today's final rule. In
todays final rule, all pneumatic device emissions are required to be estimated using emissions
factors.
Discussion of measurement options:
1) Engineering Calculations, Metering where available.
There is a paucity of data concerning the accuracy of using engineering estimates to estimate fuel
consumption. In addition, there do not appear to be standard methods available for the
determination of variables such as engine load factor. For these reasons, WCI feels that
engineering approaches to the quantification of fuel consumption are not sufficiently accurate for
this important emission source.
2) Metering required for significant fraction of fuel use, remainder by engineering calculations.
Metering of field gas consumption combined with periodic sampling of gas composition is the
most accurate and desirable manner to quantify these CO2 combustion emissions. However,
because it is likely that most field gas use is not metered, requiring the installation of new meters
at all sources may be excessively burdensome, especially considering that these units are often
relatively small and geographically dispersed among wellhead. To decrease this burden and yet
obtain data that is reasonably accurate in the aggregate, it is appropriate to require metering for
1351
only a significant fraction of the total field gas use, and estimate the remainder by engineering
calculations.
Two options were considered for setting thresholds that would trigger the metering requirement.
Both are based on the rated heat input capacity or horsepower of the combustion equipment and
both allow for a phase-in of the metering requirement, with higher thresholds the first year of
rule implementation and lower thresholds thereafter.
(a) Metering thresholds based on capacity and device type.
In this approach, the requirement to meter field gas use would be determined at the unit level,
with different thresholds depending on the type of device (compressors, heater-treaters, heaters
and additional equipment). For all devices below the threshold, fuel consumption could be
determined using an engineering approach. The primary disadvantage of this approach is that it
discourages common-pipe metering of installations having a large number of small sources.
(b) Metering threshold based on capacity of installation-level field gas combustion equipment. In
this approach, the requirement to meter field gas use is based on the installation-level sum of
heat input capacity of the field gas combustion units. This approach would tend to encourage
common-pipe metering, and could potentially cover more field gas use while requiring meters at
fewer installations.
Response: EPA disagrees with the comment. Please see the response to comment EPA-HQOAR-2009-0923-0582-17.
Subpart W should include emissions calculation methodologies and fuel use measurement and
analysis requirements specifically designed for the onshore production segment, taking into
account the type and size of combustion units, their dispersal among individual wellhead sites,
and the common availability of gas composition data obtained for production purposes.
WCI recommends Option 2b listed above. WCI recommends use of the Subpart C Tier 3
calculation method (measured fuel carbon content) but with metering and fuel analysis
requirements modified to be more feasible for oil and gas production installations. To ensure that
a significant fraction of field gas use is metered, while not requiring the installation of meters at
installations with only minor amounts of field gas use, we recommend setting an installationlevel threshold for the metering requirement. For the installations falling below the threshold, we
recommend that field gas consumption be estimated using an engineering approach such as that
given in the API Compendium of Greenhouse Gas Emission Methodologies for the Oil and
Natural Gas Industry (2009; page 4-5).
We think that an installation-level threshold is preferable to individual thresholds for each type
1352
of combustion equipment (engines, heaters, etc.) because common-pipe metering will be the
most practical approach for most operators. We recommend that this threshold be set at a rated
fuel use capacity of 525 MBtu/hr for the total of all field gas combustion equipment at an
installation. To allow time for installation of all the new meters required, we recommend a
phased approach, with the threshold set at 1675 MBtu/hr for the first year that the monitoring
rule is in effect. The higher threshold corresponds to about 200 hp and 800 metric tons CO2e/yr,
and the lower threshold corresponds to about 62 hp and 250 metric tons CO2e/yr.
For the carbon content values to be used in the calculation methodology, we recommend that
field gas composition be determined twice per year. When sufficient data are available to
evaluate variation with time for individual wells and across wells within a geological formation,
this requirement could potentially be modified to reduce the frequency of analysis or the number
of wells per formation that must be analyzed.
Response: EPA considered the points raised above but has determined that the cost of metering
fuel gas consumption in onshore production at an equipment type level or installation level
results in undue reporting burden. In todays final rule, EPA is including monitoring methods for
onshore petroleum and natural gas production that will sufficiently characterize combustion
emissions to inform policy while balancing reporting burden. Onshore production portable and
stationary combustion must follow the monitoring methods in subpart W. EPA does not agree
with the commenter on the composition requirements for onshore production. Please see Section
II.E of the preamble to todays final rule for EPAs changes regarding gas composition sampling
requirements. Onshore production operations typically know the composition of their produced
natural gas (or field gas). The incremental accuracy gained from sampling over existing
composition data does not justify the incremental cost burden associated with sampling.
WCI recommends that the following paragraph be added to SECTION 98.233:
(aa) Field gas combustion. For combustion units that combust field gas, you must comply with
following requirements:
(i) Measure the higher heating value of the field gas annually.
(ii) If the measured higher heating value is equal to or greater than 970 Btu/scf and less than
1,100 Btu/scf, then calculate the CO2, CH4, and N2O emissions using the methods in subpart C of
this part (General Stationary Combustion Sources) following the methods required for pipeline
quality natural gas.
(iii) If the measured higher heating value is less than 970 Btu/scf or greater than 1,100 Btu/scf,
then calculate the CO2, CH4, and N2O emissions using either the Tier 3 or Tier 4 methodology in
subpart C of this part (General Stationary Combustion Sources).
(iv) If the maximum rated heat input capacity of all combustion equipment located at your
facility is less than or equal to 1675 MBtu/hour for the first reporting year and less than or equal
1353
to 525 MBtu/hour in all subsequent reporting years, then fuel use may be determined by
engineering calculation based on rated heat input capacity or horsepower and load factor.
Response: EPA disagrees with the commenter. See the response to comment EPA-HQ-OAR
2009-0923-0582-17.

Section 98.233 Calculating GHG emissions
Efforts to reduce lost gas have been a driver for companies to implement component
replacement/repair programs. Management of this program requires that pre- and post-leak
measurement readings be taken to justify Operation and Maintenance (O&M) costs and confirm
fugitive emission reductions. EPA should consider that when additional direct measurements of
emissions for any component are available, the operating company could optionally use multiple
measurement results obtained during the reporting year to calculate annual emissions for the
component surveyed per the companys other reporting procedures for the prioritization of
related leak reduction work.
Response: In todays final rule, EPA allows for the adjustment of leak estimates based on
multiple surveys. Please see Section II.D of the preamble to todays final rule for further details.
EPA disagrees with the commenter on the use of leak measurements to estimate emissions. EPA
has determined that providing monitoring methods to conduct leak measurement of a sample of
leaking components and applying it to the entire facility is not practical to define in todays final
rule or necessary to inform future policy. Furthermore, such requirements would put undue
verification burden on EPA. Hence, EPA requires the use of leaker emissions factors when leaks
are found.

Proposed Rule Section 98.233 Calculating GHG emissions.El Paso encourages the EPA to
allow annual emissions to be optionally based on all direct measurement results that may be
available during a calendar year for any emission source.
Response: Please see EPAs response to comment EPA-HQ-OAR-2009-0923-1011-36.
1354

Natural Gas Driven Pneumatic Pump Venting.
As proposed, Subpart W requires oil and gas operators to calculate emissions from natural gas
driven pneumatic pump venting using an engineering estimation based on manufacturer model
emissions per unit volume and volume pumped. 40 C.F.R. 98.233(c), 75 Fed. Reg. at 18638.
However, the required manufacturer's data is not consistently available for this type of source.
Availability of such information varies based on the age of the pump and its manufacturer, and
for some older pumps, data may not be available for sufficiently "similar" pump models pursuant
to proposed 40 C.F.R. 98.233(c)(2). Additionally, while IOGA-WV appreciates that the agency
has authorized operators to utilize data for a "similar pump model, size and operational
characteristics to estimate emissions" where manufacturer data for a specific pump is not
available, we note that this creates a highly subjective standard that may vary depending upon
the interpretation of the reviewer. IOGA-WV therefore invites USEPA to provide some
additional clarification or guidelines with regard to what the agency will deem acceptable under
these provisions. Finally, IOGA-WV suggests that, in those cases when both manufacturer data
and "similar" pump data is unavailable, USEPA authorize operators to estimate emissions based
on industry knowledge and operational history, as long as the basis for such an estimate is
properly documented in company records.
Response: Please see EPAs response to EPA-HQ-OAR-2009-0923-1060-28 for further details.
Below represents challenges that specific proposed monitoring methodologies pose for many
operators.
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
Response: For all pneumatic devices and pneumatic pumps, in todays final rule, EPA is
requiring the use of emissions factors. Hence, the reporters only need to count the number of
devices to estimate emissions. See the response to comment EPA-HQ-OAR-2009-0923-1060-28.
The definitions of continuous and intermittent bleed pneumatic devices are in 40 CFR part 98.6.
For the comments on well venting for liquids unloading, EPA recognizes that the determination
of gas vented during well unloading is a difficult measurement. EPA considered the methods
that Natural Gas STAR partners used in evaluating these emissions and determined that a flow
meter on the gas vent line off a gas/liquid separator is the only practical way to measure these
emissions. EPA recognizes that the flowback is irregular, and an orifice meter recording may
therefore be irregular. Nevertheless, interpreting this meter recording will need some skill where
an average flow rate will have to be determined using engineering judgment. EPA disagrees
with the commenter that monitoring the time to vent a well during liquids unloading is difficult.
For a description of how operating personnel would collect and record the time to vent a well
during liquids unloading, please see the response to comment EPA-HQ-OAR-2009-0923-1305
46.
For the comments on gas well venting during conventional well completions and workovers
(referred to as gas well venting during completions and workovers without hydraulic
fracturing),EPA does not have sufficient information from the comment to determine why the
commenter would need meters for conventional completions. Wells are tested after completion
to determine their flow rates and producing well flow rates have to be reported to States and are
often required for royalty and tax payments. EPA requires the use of this existing data to
estimate emissions and does not see the need for any new metering.
In regards to unconventional well completions and workovers (referred to as gas well venting
during completions and workovers from hydraulic fracturing), the reporter may use their most
recent gas composition based on available sample analysis of the field. EPA has estimated that
1366
this should be manageable for most reporters. Please see response to EPA-HQ-OAR-2009-0923
1151-97 for further details on the estimated number of samples. Also, in case the reporter is not
able to conduct the required monitoring, under certain conditions, EPA is allowing for the use of
BAMM. Please see Section II.E of the preamble to todays final rule for further details.
For the comment requesting clarification on gathering pipelines, EPA has removed this source
from todays final rule. See Section II.E of the preamble to todays final rule for further details.
In todays final rule, emissions from compressors in onshore petroleum and natural gas
production need to be estimated using emissions factor and activity count. This resolves the issue
of measuring compressor emissions in different modes of operation for onshore production.
Todays final rule allows reporters to use existing GOR data; please see Sections II.D of the
preamble to todays final rule for further details. EPA has removed the reporting requirements
for CBM produced water emissions from todays final rule. For further details, please also see
Section II.E of the preamble to todays final rule and response to EPA-HQ-OAR-2009-0923
1151-129.
Regarding concerns with using flow meters for acid gas, in todays final rule EPA allows the use
of engineering estimation to determine flow rate to an Acid Gas Removal unit under certain
conditions. See EPAs response to EPA-HQ-OAR-2009-0923-1024-26 for further details. This
should avoid any issues with metering where it is not possible.
CO2 in hydrocarbon liquids (mostly crude oil and condensate) does not pass on into todays final
products (propane, butane, gasoline, etc.), hence it is a reasonable assumption that the CO2
entrained finally goes to the atmosphere. In todays final rule, EPA has clarified that the source
applies only to EOR operations. Further, sampling has been reduced to annual. This should
significantly reduce the burden to commenter on this source.
EPA has removed reporting requirements for produced water dissolved CO2 from todays final
rule which will address the commenters concerns with this source type. See responses EPA-HQ
OAR-2009-0923-1151-129.
Regarding the comment on equipment leak emissions sources, EPA does not have sufficient
details in the comment on why certain sites are not accessible for leak detection. The
accessibility that EPA has referred to in its documents relates to the physical reach of the
equipment for an operator to conduct leak detection using detection devices that need physical
contact. In fact, an IR optical imaging device or IR laser pointing device resolves this issue on
equipment access.
In regards to concentration of GHG in gas sent to flares, please see response to EPA-HQ-OAR
Please see the response to EPA-HQ-OAR-2009-0923-1060-27 for a response to the commenters
questions regarding stationary combustion reporting.
The commenter has requested BAMM for several of the sources. EPA in todays final rule is
allowing BAMM for certain source types for specified periods of time. See Section II.F of the
preamble to todays final rule for further details.
1367

Commenter: Rick Hyndman:
Pre-amble-page 78, "Under this approach, if a specific emission source is found not leaking in
the initial survey but leaking in subsequent surveys, emissions would be quantified from the date
of the first survey where a leak was detected forward through the time when the leak is fixed, or
the end of the year, whichever is first. Similarly, if an emissions source is found to be leaking in
the initial survey, emissions would be quantified from the date of that survey through to when
the leak is repaired, or the end of the year, whichever is first. Under this approach, emissions
would reflect leak reductions as determined by repairs and follow-up detection surveys.
EPA seeks comment on whether this alternative approach better estimates annual facility
emissions without resulting in additional reporting burden to the facilities. Further, we seek
comment on whether, if implemented, multiple surveys should be optional or required for
owners or operators."
- CAPP does not have a concern with the outlined methodology for assessing emissions from
fugitive equipment leaks based on the timing of the survey and subsequent leak repairs.
However, we do not believe that multiple emission surveys per year should be mandatory.
Response: In todays final rule, EPA requires at least one mandatory leak detection survey, but
allows for multiple leak detection surveys as long as they are comprehensive. In addition, EPA
has provided for the adjustment of emissions based on multiple survey results. Please see Section
II.D of the preamble to todays final rule for more details.

98.233 (d):
CAPP recommends that the use of a mass balance around the plant or amine unit be an allowed
methodology under 98.233 (d) in lieu of the requirement to directly meter V2. CAPP also
recommends the redefinition of the variable presented in the formula to:
- V1: Metered annual inlet raw gas volume into plant or into amine unit
- %Vol1: Volume weighted CO2 content of natural gas or inlet raw gas
1368
- %Vol2: Volume weighted CO2 content of sweet sales gas
In addition CAPP believes the requirement for quarterly sampling is not necessary and that the
EPA should allow for annual sampling instead.
Response: Please see EPAs response to comment EPA-HQ-OAR-2009-0923-1024-26.

CAPP recommends that in 98.236(c)(19) the venting and fugitive equipment leaks be reported
separately. This will provide a more representative picture of the contribution of fugitive
equipment leaks to the overall GHG inventory. CAPP would also like to inform the EPA that
under 98.236(c)(19)(i) the industry standard is not to prepare actual component counts as this
would require effort which is not commensurate with the benefit provided. Instead facilities
generally use an estimate based on the count of individual process units and typical component
counts for each type of process.
CAPP also seeks clarification on the sampling location the EPA would like used for
98.236(c)(19)(ii) CH4 and CO2 in produced natural gas for onshore petroleum and natural gas
production.
Response: EPA agrees with the commenter. Todays final rule requires reporters to report their
equipment leaks and vented emissions separately for each respective source. Also, todays final
rule allows onshore production reporters to quantify equipment leak emissions by counting major
pieces of equipment. For clarification on the sampling requirements for CH4 and CO2 in
produced natural gas for onshore petroleum and natural gas production, please see Section II.E of
the preamble to todays final rule for further details.
OOC Comment: To prevent confusion, the rule should clearly state that the only subsection from
98.233 that applies to offshore platforms is 98.233(s).
Response: EPA disagrees with the commenter. EPA has reviewed Sections 98.233 and
98.233(s) of todays final rule and has determined that the specific reporting requirements for
offshore operators outlined in Section 98.233 are clear and appropriate
1369

Direct measurement should not be required. Anadarko feels that direct measurement is
not appropriate for any sources for the purposes of GHG emission inventories, consistent
with other emission inventory practices. Although EPA has reduced direct measurement
requirements in the current proposal for Subpart W, EPA continues to employ methods that
ignore longstanding and accepted international protocols for emission inventory calculations.
Further, the provision for the use of best available monitoring methods stated in Section
98.3(d)(1) is exclusive to the subparts that were finalized in 2009 and were implemented
January 1, 2010. Due to the extensive and comprehensive nature of the monitoring required
by Subpart W that is not currently being conducted by this affected industrial category, we
request that EPA incorporate an amendment to Section 98.3(d)(1) that allows those subject to
any subpart that will be implemented January 1, 2011, to use BAMM for emissions for
calendar year 2011.
Response: EPA disagrees with the commenter that direct measurement is not appropriate for
any sources. Please see the response to comment EMAIL-0002-9 (comment also located in
rulemaking memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923) for a
detailed response. In addition, EPA agrees that use of best available monitoring methods should
be allowed for some sources; please see the response to EPA-HQ-OAR-2009-0923-1151-64.
Additionally, EPA disagrees that use of best available monitoring methods will be necessary for
the entire first year for all reporters. For more information, please see the Section II.F of the

Equations should be reviewed for accuracy, clarity and engineering unit consistency: Several
comments included in this document (e.g., for blowdown vents) note errors or lack of clarity
(e.g., engineering units not provided) related to equations in the Proposed Rule. While a few
specific recommendations are provided in these comments, INGAA recommends that EPA
closely review all equations for accuracy and consistency (e.g., reference conditions, consistent
engineering units for related data, proper application of molar/volume ratio for methane and CO2
content of natural gas), as it is evident that there are additional errors such as engineering units
not being provided for all equations. This review is particularly important for equations that are
frequently referenced, such as Equation W-20 and Equation W-21 in 98.233(t).
Response: EPA appreciates this comment and has thoroughly reviewed all equations and
methodologies in todays final rule for accuracy and consistency.
1370

Equation time basis should be based on operating hours: In many cases, time is (or should be) a
variable in the emission calculation. In some cases, it appears that EPA presumes 8,760 hours of
operation or source service. INGAA recommends that calculations that assume continuous
operation be revised to allow the relevant annual operating hours or time in service to be used to
calculate annual emissions.
Response: EPA agrees with the commenter, and has allowed operational hours to be used with
select sources (e.g. reciprocating compressor rod packing venting), however, has determined that
requiring reporters to track the operational time for equipment leak sources would impose an
unjustified burden on reporters. Consequently, EPA assumes when an equipment leak is
detected during an equipment leak detection survey that it has been leaking the entire reporting
period unless multiple leak detection survey are performed within the reporting period. Please
see Section II.E of the preamble to todays final rule for how reporters may adjust their
emissions with multiple leak detection surveys.

Direct measurement should not be required.
Although EPA has reduced some direct measurement requirements from the first Subpart W
proposal, EPA still chooses to ignore longstanding and accepted international protocols for
emissions calculations that other countries have adopted and currently use to calculate their oil
and gas sector emissions. This includes Australia NGERS, EU-ETS, and Alberta for which all
emissions are based on the API Compendium-1P1ECA standards. EPA is assuming that better
data will result from documenting and calculating emissions from every single small component
related to oil and gas production instead of using carefully documented emissions factors applied
across larger units. This procedure ignores the rules of probability and statistics and rejects
average values in favor of counting every bean. There is no reason to believe that this
burdensome method of data collection will produce an emissions number of greater value or
validity than other accepted and proven techniques. The proposed rule is considerably more
burdensome for onshore facilities than offshore facilities where MMS GOADS reporting
standards arc acceptable.
Response: Please see EPAs response to comment EMAIL-0002-9 (comment also located in
1371

98.233(a) Natural gas pneumatic high bleed device venting
In order to reduce burden, ConocoPhillips requests the addition of a simplified approach similar
to the one suggested for estimating number of fugitive components. i.e. number of pneumatic
controllers can be estimated as follows:
1. Require a pneumatic controller count (both types, high bleed and low bleed) for a small
number (20) of each type of major equipment (typically controllers are associated with
separators, glycol dehydrators, compressors, etc). Inventory the number of each type of major
equipment deployed in the reporting unit/area. Multiply the type of pneumatic controller counts
by the number of major equipment to arrive at a total pneumatic controller count for the major
equipment.
2. Sum the major equipment and site pneumatic controller counts for each reporting area to
arrive at the total pneumatic controller count for the reporting area and then apply the
methodology specified in the rule (98.233 (a) & (b)) to arrive at a emission estimate for the
reporting unit/area.
Sum the emission estimates for all reporting units/areas within an identified basin for purposes of
annual reporting.
Response: EPA now requires the use of emissions factors for all pneumatic devices in todays
final rule. Hence, the reporters only need to count the number of devices to estimate emissions.
In regards to estimating pneumatic device counts, please see response to EPA-HQ-OAR-2009
0923-0582-23 Please see Section II.E and Section II.F of the preamble to todays final rule for
further details.

98.233(b) Natural gas pneumatic low bleed device venting
ConocoPhillips requests the addition of a simplified approach for estimating the number of low
bleed controllers. An approach similar to the one discussed above for high bleed controllers
should be allowed. Most of these devices actuate only a few times per day. Probably 99% of our
pneumatics are venting << 1% of the time. In order to reduce overestimating emissions, EPA
should give us the option to use a statistical approach (i.e. track 2% of sites each year or
something similar) to estimate the time factor. The calculation approach would be documented in
1372
the GHG Monitoring Plan.

0923-0582-23. See Section II.E and Section II.F of the preamble to todays final rule for further
details.
Commenter:
Source Type:
NG pneumatic high bleed device venting
Regulatory reference for calculation/ monitoring requirements
40 CFR 98.233(a)
1. Natural gas driven pneumatic device continuous bleed rate volume at standard conditions in
cubic feet per minute, as provided by the manufacturer.
2. Amount of time in minutes that the pneumatic device has been operational through the
reporting period.
Comment:
Tracking natural gas bleed rate during normal operation for every high bleed device will be
difficult to compile for the thousands of devices owned by Yates Petroleum Corporation. Yates
asks the EPA to propose a reasonable high bleed emission factor to use in place of
manufacturer's data if manufacturers data is unavailable. Further, It is unclear what the EPA
means by "continuous." If it is a "continuous" bleed device, why is there a time element in the
equation? Lastly, it is unclear at what point is the unit considered not in operation and therefore
exempt.
Source Type:
NG pneumatic low bleed device venting
40 CFR 98.233(b)
1. Total number of natural gas pneumatic low bleed devices.
2. For onshore petroleum and natural gas production facilities, concentration of GHG i, CH4 or
CO2, in produced natural gas; for facilities listed in 98.230(a)(3) through (a)(8), GHGi equals
1.
Comment:
1373
It is unclear what the EPA means by "continuous." It is unclear at what point is the unit
considered not in operation and therefore exempt. A general note regarding the treatment of high
and low-bleed devices: Yates recommends that EPA harmonize treatment of these sources with
the API Compendium.
Source Type:
NG driven pneumatic pump venting
40 CFR 98.233(c)
1. Natural gas driven pneumatic pump gas emission in emission per volume of liquid pumped
at operating pressure in scf/gallon at standard conditions, as provided by the manufacturer.
2. Volume of liquid pumped annually in gallons/year.
Comments:
It is unclear at what point is the unit considered not in operation and therefore exempt. It is
impractical to keep the records that the EPA is proposing. It does seem from supporting
documentation that it is not expected that these sources are a large emitter of GHGs. Therefore,
utilizing man-hours to track and maintain a library of emission data from these pumps (of which
YPC operates thousands) is overly burdensome to report a relatively small amount of GHG
emissions. This data is kept by well-site, not by individual pumps. YPC proposes allowing a sitewide volume of liquid pumped.
Response: Upon further analysis and review, EPA has determined that good activity data
(pneumatic device count) and acceptable emissions estimates from pneumatic devices can be
achieved using emissions factors for high bleed, low bleed and intermittent bleed pneumatic
controllers. Therefore todays final rule does not require the reporters to consult instrument
vendors for bleed rates or maintain operational records of the time a pneumatic controller is in
service. The emission factors for bleeding pneumatic devices are taken from the third edition of
the API Compendium. In the context of this rule, the term continuous bleed refers to a
continuous flow of pneumatic supply gas to the process measurement device (e.g. level control,
temperature control, pressure control) where the supply gas pressure is modulated by the process
condition, and then flows to the valve controller where the signal is compared with the process
set-point to adjust gas pressure in the valve actuator.
EPA now requires the use of emissions factors for all pneumatic devices in todays final rule.
Hence, the reporters only need to count the number of devices to estimate emissions. In regards
to estimating pneumatic device counts, please see response to EPA-HQ-OAR-2009-0923-0582
23. Please see Section II.E and Section II.F of the preamble to todays final rule for further
details.
Commenter:
1374

Comment on 98.233:
Equations should be removed from the proposed rule and placed in a guidance document. This
will allow for corrections and updates whenever necessary without prompting a rule change.
Response: EPA disagrees with this comment. In order to change any equations, emissions
factors etc., EPA must re-propose the rule and initiate the appropriate public comment process.
There is no system by which EPA could simply update equations and factors without prompting
a rule change. Hence, EPA has retained the equations in the subpart W rule text.
Commenter:
EPA seeks comment on whether there are additional or alternative software packages to E&P
Tank and GlyCalc that should be required to be used to calculate emissions. (page 65)
MidAmerican is not aware of any other specific software packages that should be considered at
this time to calculate emissions. However, alternatives should be considered as they become
available as long as they have the necessary functionality to appropriately calculate and capture
the data.
Response: EPA now allows reporters to use any software package to calculate onshore
production tank emissions that meet the criteria specific in todays final rule.
Provide streamlined emission-estimating methods for upstream production sites with limited
GHG emissions. For these sites, the API Compendium provides a reliable, accurate, and
streamlined method of estimating emissions.
Response: EPA disagrees with the commenter. Most production sites with limited GHG
emissions in the U.S. will likely not cross the 25,000 metric ton CO2e threshold, and those
basin-operators that do cross the threshold have been estimated to emit 85% of the emissions
from the petroleum and natural gas sector.
1375

Source Category Comments and Alternative Emission Estimation Methodologies
API proposes alternative estimation methodologies for most of the source types applicable to the
Onshore Petroleum and Gas Production Sector (OGPG). These alternative methodologies are
warranted to reduce the man power and cost burden and to address errors in EPAs assessment of
specific sources and methodologies. Also, API recommends that some of the source types be
deleted from the final rule because of insignificant GHG emissions associated with them
(reasons/examples are provided below for relevant source type discussions).
A. Section 98.233(a) Natural gas pneumatic high bleed device venting
Onshore Production Source Category: Natural gas pneumatic high bleed device venting The
proposed rule requires gathering manufacturer data for each unique device model and
determining the amount of time each device is operational during the reporting year. API
estimates that some 673,000 locations would need to be visited to gather this information,
requiring an estimate 667 man-years (based on 2000 hours per year).
The inventorying of pneumatic controllers should be simplified by a method similar to the one
suggested for estimating number of fugitive components (provide in item N below). API
suggests that the number of pneumatic controllers will be estimated as follows:
1. Require a pneumatic controller count (by high bleed and low bleed types) for a small number
(5) of each type of major equipment (typically controllers are associated with separators, glycol
dehydrators, compressors, etc) within the Sub-basin entity. Inventory the number of each type of
major equipment deployed in the Sub-basin entity. Multiply the type of pneumatic controller
counts by the number of major equipment to arrive at a total pneumatic controller count for the
major equipment.
2. Sum the major equipment and site pneumatic controller counts for each Sub-basin entity to
arrive at the total pneumatic controller count for the Sub-basin entity and then apply the
methodology specified in the rule (Section 98.233 (a) & (b)) to arrive at a emission estimate for
the Sub-basin entity.
3. Sum the emission estimates for all Sub-basins within an identified Basin entity for purposes of
annual reporting.
Per equation W-1, the rule requires calculation of emissions for each pneumatic controller by
keeping record of the amount of time the pneumatic controller has been operational through the
reporting period. API requests flexibility in determining the operational time of high bleed
pneumatic controllers:
1. Where the operational time of high bleed controllers is far less than 8760 hours per year, (as a
matter of fact it could be much less than 1% of the time), API requests the option of estimating
1376
operational time based on statistical samples/periodic surveys of the pneumatic controllers in the
Sub-basin entity so that a Sub-basin default operational hours/year for each Basin entity can be
applied.
2. Where a Sub-basin entity is known to have operated continuously or nearly continuously for
the year, API requests the flexibility of estimating emissions based on 8760 hours of operation.
3. Where a Sub-basin entity operates for a portion of the year, API requests the option of
adjusting the number of days of operation to correspond to full days that the pneumatic controller
and associated equipment at the Sub-basin were in operation
Also, equation W-1 refers to paragraph Section 98.233 (u) to calculate GHG emissions based on
the composition of the produced natural gas used to operate the controllers. See Section III
comment #39 above for APIs suggested revisions to sampling frequency.
0923-0582-23. See Sections II.E and II.F of the preamble to todays final rule for further details.
In regards to produced natural gas composition, see response to EPA-HQ-OAR-2009-0923

Where a high bleed pneumatic controller has been retrofitted, such that the manufacturers
information is no longer applicable; API proposes that an emission factor post-retrofit be applied,
if available, otherwise the low bleed emission factor should be applied to these pneumatic
controllers.
Where a manufacturer or model number can not be determined, due to age of the pneumatic
controller or inaccessible information, API requests the use of a default emission factor from the
API Compendium.
0923-0582-23. See Sections II.E and II.F of the preamble to todays final rule for further details.
Where a high bleed device retrofitted to reduce bleed rate is bleeding at or below the low bleed
device bleed rate as defined in todays final rule, the reporter should use the low bleed device
emissions factor.
1377

Section 98.233(b) Natural gas pneumatic low bleed device venting
Onshore Production Source Category: Natural gas pneumatic low bleed device venting
API proposes the same phased sampling approach and provisions for retrofits or inaccessible
information for low bleed pneumatic controllers as described for high bleed pneumatic
controllers.
Equation W-2 for low bleed pneumatic controllers incorporates factors for continuous annual
operation (24 365), rather than using a controller-specific time factor, as provided in Equation
W-1. API proposes the same options for determining the operational time of low bleed
pneumatic controllers as described above for high bleed pneumatic controllers
See Section II.E of the preamble to todays final rule for further details. In regards to retrofits to
high bleed devices, please see response to EPA-HQ-OAR-2009-0923-1151-111.
Section 98.233(c) Natural gas driven pneumatic pump venting.
Onshore Production Source Category: Natural gas driven pneumatic pump venting
The proposed rule requires gathering manufacturer data for each unique pump model per unit
volume of liquid circulation rate at pump speeds and operating pressures, and the amount of
liquid pumped annually for individual pumps. Pneumatic pumps are used for the injection of
small amounts (i.e., several ounces of fluids over several hours). In addition, the pumps are
installed on sumps, which are typically installed on every wellhead and manifold. One API
company reports there are approximately 1,000 such pumps spread out over a 600 square mile
area with approximately 5 people responsible for maintaining the equipment and visiting a site
about every two weeks. On a national level, API estimates that this requirement applies to
approximately 520,000 pneumatic pumps and would require 260 man years (based on 2000
hours per year) to gather the required information for the first year. Keeping a log for each pump
is impractical. API recommends engineering estimates be used for volume of liquid pumped
annually.
Three primary types of pneumatic pumps are used in oil and gas production operations:
- Pneumatic pumps used in association with dehydration units, generally referred to as Kimray
1378
pumps. API requests that Section 98.233(b) specifically exclude reporting emissions for
pneumatic pumps associated with glycol dehydration units. Emissions from these pumps are
already included in the GLYCalc modeling required for the dehydration units and will be
captured in emissions reported under Section 98.233(e).
- Piston pumps; and
- Dual diaphragm low pressure pumps.
Operators do not log how much liquid each pump handles, so determining emissions based on
the volume handled for each pump would be an excessive burden, as noted above. API requests
the following alternatives.
For the piston pumps, company records of the types of pumps and pump curves should be
allowed to determine the average amount of gas released per volume of liquid pumped.
Emissions can then be quantified for the Sub-basin entity by applying company records of the
amount of chemicals or methanol purchased/injected to the average gas usage rates.
For diaphragm pumps, API requests the use of a representative sampling of pumps for the Subbasin entity to determining the number of strokes per minute and an estimated operational
duration for the year. The Sub-basin entity data would be combined with manufacturer
information on the amount of gas displaced per stroke to quantify the emissions for a grouping of
similar pumps.
Response: EPA now requires the use of emissions factors for pneumatic pumps in todays final
rule. Hence, the reporters only need to count the number of pumps to estimate emissions. See
section II.E of the preamble to todays final rule for further details.
EPA agrees with the commenter with regards to double counting of emissions from glycol
circulations pumps. Please see the response to EPA-HQ-OAR-2009-0923-0582-33 for further
details.
Section 98.233(a) Natural gas pneumatic high bleed device venting. The equation seems to be
geared towards continuous bleed devices. It will be extremely difficult to track the duration of
actuation for intermittent bleed devices. API requests that the definition in Section 98.7 match
what appears to be requested in Section 98.233(a), which is continuous bleed devices. Therefore,
intermittent bleed devices are not addressed.
See Section II.E of the preamble to the final rule for further details.
1379


Section 98.233(b) Natural gas pneumatic low bleed device venting. Equation W-2 provides
factors to convert between scf of CH4 and CO2 to tonnes of CO2 equivalents. Applying a molar
volume conversion of 379.3 scf/lbmole, based on industry standard conditions of 60F and 1
atm, these emission factors would be 0.0004018 for CH4 and 0.00005262 for CO2. It appears the
factors shown for Section 98.233(b) are based on standard conditions of 56F and 1 atm. API
urges EPA to adopt industry standard conditions of 60F and 1 atm.
Response: EPA has revised todays final rule, and emission factors in Tables W-1A, W-2, W-3,
W-4, W-5, W-6, and W-7 are all at standard conditions of 68F and 14.7 psia, as defined in
subpart A.
Section 98.233(b) Natural gas pneumatic low bleed device venting - Paragraph Section 98.233(b)
provides the emission calculation methodology for natural gas pneumatic low bleed device
venting. The methodology uses number of devices and the emission factors from Tables W-1,
W-3, and W-4 to calculation emissions. Paragraph Section 98.233(r) provides the population
count and emission factor emission calculation methodology for certain emissions sources
including pneumatic low bleed device venting for onshore petroleum and natural gas production
facilities (Section 98.232(c)(2)), onshore natural gas transmission compression facilities (Section
98.232(e)(6)), and underground LNG storage facilities (Section 98.232(f)(4)). Paragraph Section
98.233(r) uses the number of devices and emissions factors from Tables W-1, W-3, and W-4 for
pneumatic low bleed devices. It appears the calculation methodology in Section 98.233(b) and
Section 98.233(r) for pneumatic low bleed devices is duplicative.
17. Table W-2 is missing a Low Bleed Pneumatic Device emission factor, although this table is
referenced for the emission factor in Section 98.233(b), Equation W-2.
Response: EPA agrees with the comment, and has revised todays final rule to remove the
reference to pneumatic devices from Section 98.233(r). In todays final rule, pneumatic bleed
device or pump venting for onshore natural gas processing are not required to be reported, since
most of these devices in natural gas processing are driven by air, and may not contribute
significantly to GHG emissions.
1380

Section 98.233(c)(2) Natural gas driven pneumatic pump venting. Section 98.233(c)(1)(ii)
requires reporters to maintain a log of the amount of liquid pumped annually from individual
pumps. Keeping a log for each pump is impractical. API recommends engineering estimates be
used for volume of liquid pumped annually. Additional details are provided in Section VII.C of
this document.
Section II.E of the preamble to todays final rule for further details.

Section 98.232(c)(3) and Section 98.233(c): Pneumatic pumps are used in the onshore
production sector to pump chemicals, methanol, heat trace fluid, and other materials. An
individual well site may have no pneumatic pumps or multiple pumps in different services. API
opted to estimate the number chemical and methanol injection pumps rather than apply the value
from EPAs national inventory, which appears very low (62,253 CIPs). Assuming that there is an
average of one pneumatic pump per gas well and 0.5 pneumatic pumps per oil well, this yields a
population of about 520,000 pneumatic pumps. Note this does not include pneumatic pumps
associated with dehydrators (commonly referred to as Kimray pumps) because these emissions
are accounted for in running the GLYCalc model. The rule requires tracking the amount of liquid
pumped and the discharge pressure for each pump, and emissions calculated based on the pump
curves for the specific pump. Assuming 81% are covered by the rule, and projecting one hour per
pump to fulfill the rule requirements at $100 per hour, yields a total cost of $52 MM.
rule. Hence, the reporters only need to count the number of pumps to estimate emissions. This
should alleviate any cost concerns related to reporting emissions from pneumatic pumps. See
details.

1381

Pneumatic Control Devices
We support WCIs recommendations on emissions reporting for pneumatic control devices. We
agree that metering should be required for all high-bleed pneumatic devices and all natural gas
powered pneumatic pumps. We agree that engineering calculations using emission factors is
appropriate for low-bleed devices.
See Section II.E of the preamble to todays final rule for further details.
Noble supports Section VII of the API Subpart W comments and proposed alternative methods
as outlined in this section. - Noble generally supports API comment regarding the following
emission sources: natural gas pneumatic high bleed device venting, natural gas pneumatic low
bleed device venting, well venting for liquids unloading, gas well venting during conventional
well completions, gas well venting during conventional well workovers, reciprocating
compressor rod packing venting, dehydrator vent stacks, storage Tanks, associated gas venting
and flaring , centrifugal compressor wet seal degassing venting, coal bed methane produced
water emissions, EOR injection pump blowdown, acid gas removal vent stack, hydrocarbon
liquids dissolved CO2, produced water dissolved CO2, and fugitive emissions.
Response: Please see EPAs responses to API comments referenced below for each emission
source.
Natural gas pneumatic high bleed device venting and natural gas pneumatic low bleed device
venting:
EPA now requires the use of emissions factors for all pneumatic devices in todays final rule.
Hence, the reporters only need to count the number of devices to estimate emissions. See Section
II.E of the preamble to todays final rule for further details.

Well venting for liquids unloading:
Please see the response to comment EPA-HQ-OAR-2009-0923-1151-118
Gas well venting during well completions without hydraulic fracturing and gas well venting
during well workovers without hydraulic fracturing:
Reciprocating compressor rod packing venting:
1382
Dehydrator vent:
Storage Tanks:
Associated gas venting and flaring:
Centrifugal compressor wet seal degassing venting:
Coal bed methane produced water emissions:
EOR injection pump blowdown:
Acid gas removal vent stack:
Hydrocarbon liquids dissolved CO2:
Produced water dissolved CO2:
Equipment leak emissions:.

Natural gas driven pneumatic pump venting: We request that EPA allow an operator to use an
engineering estimate of the count of natural gas driven pneumatic pumps. Therefore, we also
request that EPA modify Section 98.233(c)(1)(ii) to allow an operator to maintain a record of the
maximum amount of liquid pumped annually from pumps on a facility-wide basis and the
emissions are to be calculated on the facility-wide basis as well rather than for each pump.
1383
should alleviate burden to report from this sources. Therefore, EPA disagrees with the
commenter to allow estimating the count of pneumatic pumps. See Section II.E of the preamble
Natural gas pneumatic high-bleed device venting
Section 98.233(a): For a reporter to count and obtain manufacturers specifications for
potentially thousands of high-bleed pneumatic devices presents an excessively costly and timeconsuming effort. Therefore, IPAMS requests that EPA allow a reporter to use an engineering
estimate of the count of high-bleed pneumatic devices in conjunction with a typical bleed rate to
calculate emissions from this source category. This engineering estimate could then be applied
on a basin-wide level.
0923-0582-23. Please see Sections II.E and II.F of the preamble to todays final rule for further
details.

Natural gas pneumatic low-bleed device venting
Section 98.233(b): For a reporter to count potentially thousands of low-bleed pneumatic devices
presents an excessively costly and time-consuming effort. Therefore, IPAMS requests that EPA
allow an operator to use an engineering estimate of the count of low-bleed pneumatic devices.
This engineering estimate could then be applied on a basin-wide level.
Response: In regards to estimating pneumatic device counts, please see response to EPA-HQ
OAR-2009-0923-0582-23. Please Section II.F of the preamble for further details.
Natural gas driven pneumatic pump venting
1384
Section 98.233(c): IPAMS requests that EPA allow an operator to use an engineering estimate of
the count of natural gas-driven pneumatic pumps. Therefore, IPAMS also requests that EPA
modify Section 98.233(c)(1)(ii) to allow an operator to maintain a record of the maximum
amount of liquid pumped annually from pumps according to and applied by appropriate size
categories.
should alleviate burden to report from this sources. Therefore, EPA disagrees with the
commenter to allow estimating the count of pneumatic pumps. See Section II.E of the preamble
Finally, the technical support document for this proposed rule makes numerous other
inapplicable assumptions about LNG facilities, equipment, and practices. First, the document
assumes that there are twice as many satellite storage facilities as accounted in the industry. The
EPA also assumes that all boiloff from the LNG in storage tanks is re-liquefied by compressors
at all facilities, which is not the case. The increase in the universe of LNG facilities, the
presumption that all facilities operate numerous compressors (instead of the LNG pumps used in
most of those instances), and the assumption that the compressors used are all of the higheremission type common to other industry segments, along with the misunderstandings noted
above regarding specialized equipment and the vigilant operations and maintenance practices in
place at these facilities, all render inapplicable the subjects and methods proposed for quantifying
emissions of greenhouse gases from LNG facilities.
Response: EPA disagrees with the commenter on LNG facilities. The listing of satellite storage
facilities is as provided in the data source available from Gas Technology Institute and listed in
the references to the rule. Hence, EPA did not estimate these facilities, rather used a published
data set. However, EPAs analysis demonstrated almost none of the satellite facilities cross the
threshold. With regards to compressors, EPA agrees that if an LNG facility does not have any
compressors then they simply do not report for that source. The commenter does not provide any
analysis or measurement data to back its assertions that compressor emissions from LNG
facilities are lower than other compressor stations. Also, LNG facilities in most cases handle
natural gas and hence are liable to equipment leaks just like in other gas handling facilities.
Hence, EPA has retained the requirements for reporting from LNG facilities.
1385

The equation seems to be geared towards continuous bleed devices. It will be extremely difficult
to track the duration of actuation for intermittent bleed devices. We request that the definition in
98.7 match what appears to be requested in 98.233(a), which is continuous bleed devices.
Therefore, intermittent bleed devices are not addressed. Retaining a record of the minutes that a
particular device is operational is not feasible
H. Section 98.233(b) Natural gas pneumatic low bleed device venting.
Equation W-2 provides factors to convert between scf of CH4 and CO2 to tonnes of CO2
equivalents. Applying a molar volume conversion of 379.3 scf/lbmole, based on industry
standard conditions of 60 degrees F and 1 atm, these emission factors would be 0.0004018 for
CH4 and 0.00005262 for CO2. It appears the factors shown for 98.233(b) are based on standard
conditions of 56F and 1 atm. EPA should adopt industry standard conditions of 60F and 1 atm.
Please see comment EPA-HQ-OAR-2009-0923-1151-30 for a response to the emissions factor
values.


Section 98.233(c)(1)(ii)
Section 98.233(c)(1)(ii) requires reporters to maintain a log of the amount of liquid pumped
annually from individual pumps. Keeping a log for each pump is impractical. BP recommends
engineering estimates be used for volume of liquid pumped annually.

1386
The proposed rule requires inventorying and gathering manufacturer data for each unique device
model, and determining the amount of time each device is operations during the reporting year.
To accomplish this, each individual site would have to be visited and the specific model number
and type of controller determined and logged.
To enable this inventory, BP suggests that it be spread over a 3 year period with 1/3 of the sites
being inventoried each year. For the first two years the inventory conducted would be applied to
the total site population.
Per equation W-1, the rule requires calculation of emissions for each pneumatic controller by
keeping record of the amount of time the pneumatic controller has been operational through the
reporting period.
- Where a sub-basin entity is known to have operated continuously or nearly continuously for the
year, BP requests the flexibility of estimating emissions based on 8760 hours of operation.
- Where a sub-basin entity operates for a portion of the year, BP requests the option of adjusting
the number of days of operation to correspond to full days that the pneumatic
Where a manufacturer or model number can not be determined, due to age of the pneumatic
controller or inaccessible information, BP requests the use of a default emission factor from the
API Compendium.
B. Section 98.233(b) Natural gas pneumatic low bleed device venting
BP proposes the same phased approach and provisions for low bleed pneumatic controllers as
described for high bleed pneumatic controllers.
C. Section 98.233(c) Natural gas driven pneumatic pump venting.
The proposed rule requires gathering manufacturer data for each unique pump model, per unit
volume of gas usage per unit of liquid pumped at pump speeds and operating pressures, and the
amount of liquid pumped annually for individual pumps. Keeping a log for each pump is not
practical and BP recommends engineering estimates be used for volume of liquid pumped
annually.
Three primary types of pneumatic pumps are used in oil and gas production operations:
- Pneumatic pumps used in association with dehydration units, generally referred to as Kimray
pumps. BP requests that 98.233(b) specifically exclude reporting emissions for pneumatic pumps
associated with glycol dehydration units. Emissions from these pumps are already included in the
GLYCalc modeling required for the dehydration units and will be captured in emissions reported
under 98.233(e).
- Piston pumps; and
1387
- Dual diaphragm low pressure pumps.

Operators do not typically track how much liquid each pump handles, so determining emissions
based on the volume handled for each pump would be an excessive burden, as noted above. BP
requests the following alternatives:
- For the piston pumps, company records of the types of pumps and pump curves should be
allowed to determine the average amount of gas released per volume of liquid pumped.
Emissions can then be quantified for the reporting unit by applying company records of the
amount of chemicals or methanol purchased/injected to the average gas usage rates.
- For diaphragm pumps, BP requests the use of a representative sampling of pumps for the
reporting unit to determining the number of strokes per minute and an estimated operational
duration for the year. The reporting unit data would be combined with manufacturer information
on the amount of gas displaced per stroke to quantify the emissions for a grouping of similar
pumps.
Response: EPA now requires the use of emissions factors for all pneumatic devices and
pneumatic pumps in todays final rule. Hence, the reporters only need to count the number of
devices to estimate emissions. In regards to estimating pneumatic device counts, please see
response to EPA-HQ-OAR-2009-0923-0582-23. This should alleviate burden to report from this
sources. Therefore, EPA disagrees with the commenter to allow a phased in approach for these
devices. Please see Sections II.E and II.F of the preamble to todays final rule for further details.
details.

Organization: Sierra Club
Commenter: Anne Harvey

In our comments on the full reporting role, we discussed how engineering estimates and
emissions factor-based approaches had often underestimated emissions, and in particular,
fugitive emissions from oil and gas facilities. Were concerned that EPA has moved too far away
from direct measurement in this proposed role in favor of a range of estimation approaches. We
encourage EPA to carefully consider whether this decision causes any loss in accuracy and
precision and whether such losses can be justified given the reporting rules focus on
comprehensive and detailed information.
Response: EPA has chosen monitoring methods to balance between emissions accuracy and
cost to industry. Using costly monitoring methods for marginal gains in emissions accuracy does
not provide much benefit in terms of policy making. Hence, although EPA has chosen emissions
1388
factors and engineering estimates for several monitoring methods, those methods will provide
sufficient clarity in terms of magnitude of emissions for any future policy considerations.
Organization: Environmental Defense Fund
Commenter: Peter Zalzal
As a general rule, emissions data derived from direct measurement is more accurate than data
derived from engineering estimates and emissions factors. In the oil and gas sector, this rings
especially true because emissions factors tend to be based on older sources of data and
considerable variability can exist between sites due to differences in operations and maintenance
conditions. For example, although two oil and gas sites may have the same number of
components and types of equipment, if the operator of one site invests considerably more
resources into maintenance and leak detection of his equipment, the overall leak rate on a percomponent basis for that site will be much less than alternative sites. Components with higher
occurrence of leaking such as compressors and equipment with less rigorous or dated emissions
factors, should therefore be required to utilize direct measurement over engineering estimates.
Response: EPA agrees with the commenter that there is some variation between facilities in
terms of equipment leaks. However, EPA concluded that requiring leak detection and the
application of leaker emissions factors will substantially mitigate that issue. Furthermore, leaks
occurring across facilities still leak at similar average rates; i.e., connector leak across different
facilities will have similar average leak rates. The impact of maintenance practices will be
reflected in the number of leaks and not as much in the average leak rate. Hence, EPA has
retained the requirements for emissions factors in todays final rule.
1389
13.2.1
WELL VENTING, COMPLETION AND UNLOADING CALCULATIONS

Measurement Options: WCI evaluated the following four options.
WCI Recommendation: USEPAs approach to calculating emissions from well venting is the
most accurate of the available methods. WCI recommends the use of USEPAs approach (option
4) with modifications to SECTION 98.233(f) and (g). WCI recommends the SECTION
98.233(h) be adopted without modification.
1390
WCI Modifications: WCI recommends the following modifications to USEPAs approach:

(1) Well venting for liquids unloading (SECTION 98.233(f))
Limit options to Calculation Method 2 where individual well unloading event emissions are
calculated.
Draft Calculation method 1 requires use of a recording flow meter only for each unique well
tubing diameter and producing horizon/formation combination in each gas producing field where
gas wells are vented to the atmosphere to expel liquids accumulated in tubing. In addition,
Calculation method 1 only requires measurement of flow in each unique combination, every
other year. Thus Calculation method 1 would not produce cap-and-trade quality data.
(2) Gas well venting during unconventional well completions and workovers (SECTION
98.233(g))
Use either Calculation Method (1) or (2), for each well completion and workover.
The proposed USEPA method requires measurement of flow rate only for each unique tubing
diameter and producing horizon/formation in each producing field every other year. This limited
sampling would not result in cap-and-trade quality data. Therefore, WCI recommends that these
methods be applied to every well completion and workover. Text should be added stating that in
situations where vented gas is captured for use (e.g., green completions) reporting emissions
would not be required. If vent gas is sent to a flare emissions would be calculated in the flare
section section (n).
Response: EPA disagrees with the commenter. The rule has proposed methodologies that
estimate emissions from sources within a level of accuracy that effectively informs future policy.
Tightening reporting methodologies for greater accuracy will come at a substantially higher cost,
which cannot be justified by the incremental accuracy gains for the purposes of this rule.
Therefore, EPA has retained the well venting monitoring methods as proposed in the April 2010
proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002).
(98.233 (g)) Many times the producer knows the choke size of wells that are venting to the
atmosphere. Accurate gas rates can be calculated knowing the upstream pressure, choke size and
downstream pressure. Would the EPA consider this as a method to calculate vented gas rates?
Response: EPA is not recommending a flow calculation across the well choke for liquids
unloading because there are two more accurate methods required in todays final rule: a vent
meter or calculations based on well dimensions and operating parameters. Choke calculations
1391
during transient pressure flow can be a complex calculation with sonic flow part of the time and
subsonic flow for all or the remainder of the unloading time. The pressure upstream of the tubing
string and choke is very uncertain with liquids being lifted inefficiently without a plunger lift and
highly variable depending on the amount of liquid accumulation at the time a well is selected for
liquid unloading. EPA has decided that the direct measurement of vent gas or the equations
based on well depth, casing diameter, tubing diameter, shut-in and sales line pressures, and
venting time will be cost-effective and adequately accurate in characterizing emissions.

Proposed Rule Section 98.233(g) Gas well venting during unconventional well completions and
workovers.As in several other recommendations related to providing reporters monitoring
flexibility, El Paso requests the third method from the Technical Support Document be included
as an alternative method in the rule, since the approach uses information commonly tracked from
completion and workover activities.
Response: EPA disagrees with the commenter. Although the option is listed in the TSD, the flow
rate during completions and workovers is often more than the flow rate during normal
production. This is because during a completion or workover there is no restriction to the flow of
the well, unlike during production wherein the choke at the wellhead controls the flow rate. Also,
EPA has already provided this option for well completions and well workovers that do not need
hydraulic fracturing. However, when a hydraulic fracture is conducted the flow rate and time for
well completions are significantly higher. Hence, EPA wants to correctly capture this
information. Finally, EPA has analyzed and determined that the samples required with well
completions and well workovers that require hydraulic fractures is manageable. For further
Section 98.233(g) Gas well venting during unconventional well completions and workovers
El Paso estimates that it would have to report emissions for approximately 3,500 unconventional
and conventional completions or workovers using the basin level reporting threshold.
EPA assumes that the majority of unconventional wells are completed with hydraulic fracturing.
The proposed rule details two different methodologies for estimating emissions from
unconventional well completions or workovers:
- Method #1 requires the measurement of flow rate during clean-up for one well completion and
1392
one well work-over in each field. From this a flow rate per minute is derived and applied to the
minutes of flow-back from each completion and work-over in a field.
- Method #2 requires monitoring across the completion choke and calculation of flow rate during
clean-up for one well completion and one well work-over in each field. From this a flow rate per
minute is derived and applied to the minutes of flow-back from each completion and work-over
in a field.
The TSD considered a third method for determining emissions from completions and workovers.
Method 3 is based on the daily gas production rate, the amount of venting time, and the
composition of CH4 and CO2 in the produced gas. This approach uses information commonly
tracked from completion and workover activities and should be included as an alternative
method in the rule.
Response: EPA disagrees with this comment. With regards to the third method in the TSD
referenced for calculating emissions from completions and workovers, please see the response to
EPA-HQ-OAR-2009-0923-1011-11.
Well Venting for Liquids Unloading
With regard to well venting from liquids unloading, Subpart W proposes to require oil and gas
operators to monitor emissions using one of the following alternatives: (1) engineering
estimation based on a field-specific emissions factor multiplied by the number of events or (2)
direct measurement using a flow-metered emissions factor multiplied by the number of events.
40 C.F.R. 98.233(1), 75 Fed. Reg. at 18639. Given the large number of wells that a single oil
and gas operator can control, however, installing a flow meter on each vent line is impractical.
The necessary maintenance of the meter and the time required to monitor and subsequently
compile the information into useful data is extensive for the minimal amount of emissions that
occur during this process. Moreover, the engineering estimation will be well-specific and will
also require considerable time and expense to complete for each well. Particularly if USEPA
retains its basin-level definition of an onshore production facility, these requirements could apply
to hundreds or thousands of wells within a particular basin, thereby creating an enormous burden
on operators. Because this process is not a significant source of GHG emissions, and because the
burdens associated with calculating these emissions outweigh the environmental benefits
associated with quantifying the minimal emissions associated with this process, IOGA-WV urges
the elimination of this section from the final rule.
Response: EPA disagrees with the commenter on the methodology issue for calculating
emissions from well venting from liquids unloading or the exclusion of the source from
reporting. EPA provided two alternatives in the proposal to estimate emissions from this source.
The first is an engineering estimation method that applies to all wells. In this methodology the
only two unknown parameters are number of times the wells are unloaded and the time required
for liquids unloading. This parameter can be tracked using a log by the operator who opens the
1393
well to the atmosphere where unloading is manually conducted. In the case of plunger lifts, the
timing cycle of the plunger can be tracked to determine hours required for unloading. The well
depth, shut-in pressure, and casing/ tubing diameter are all known data for operators. Hence
conducting this calculation is partly desktop and not resource intensive, since no new equipment
of monitoring methods have to be put in place.
The second alternative provided in the proposal was to estimate unloading emissions using
meters on each unique well tubing diameter and producing horizon/ formation combination in
gas producing field where liquids unloading occur. EPA wishes to clarify that the intent is to
meter representative wells in a field and not meter every single well. The representative sample
emissions should then be extrapolated to all wells that the sampled well represents. This is the
least burdensome way of estimating emissions that are reasonably accurate.
Finally, EPA would like to highlight that the emissions from well liquids unloading are 34
percent of the total emissions from onshore production. Hence due to its significance EPA did
not consider it appropriate to exclude the source from reporting. In todays final rule, EPA has
retained the proposal methodology for well liquids unloading.
98.233 (g):
CAPP would like clarification on 98.233 (g)(i) and requests that the EPA provide a definition of
a "gas producing field", as well as guidance on how to distinguish fields. Additionally under
98.233(g)(i)(A) CAPP requests that the EPA provides guidance on how to calculate the
emissions from a well completion within the field.
Response: EPA agrees with the commenter. In todays final rule, EPA has provided a definition
of a gas well that should be used to apply monitoring methods. Also, EPA has provided
calculation methods for estimating emissions using pressure drop readings across a choke in
todays final rule.
98.233 (h):
When reviewing 98.233(h) CAPP noticed that equation W-9 is incorrect as it references Daily
Gas Production Rate. This equation would be correct if the back-pressure remained constant,
however since it fluctuates, equation W-9 should be changed to incorporate the correlation
between venting and well back-pressure.
Response: EPA agrees with the commenter and understands that the back-pressure varies over
time. Hence in todays final rule, the equation requires the calculation of the production rate
1394
using an annual average that will balance out the changes in the back-pressure over a year. To
keep the calculation method simple, EPA requires the use of an average value.
It is unclear if valve operators are intended to be included in vented emissions as they appear to
be excluded from pneumatic devices. EPA should clarify whether valve operators are to be
included as a vented emission or as part of the pneumatic device with which it is associated
Response: Upon further analysis and review, EPA has determined in todays final rule that high
bleed, low bleed and intermittent bleed gas pneumatic controllers shall be reported using
component counts and population emission factors. These emission factors cover each natural
gas powered, automated process control loop such as level, temperature and pressure controllers.
Manual, natural gas powered isolation or shut-off valves and non-venting pneumatic pressure
regulators are not included in todays final rule.
Well Completion Emission Factor
For well completion venting and well work-over venting EPA proposes the development of a
field-specific emission factor either by direct measurement of flow rate or hydrocarbons (not
methane) using a meter or by an engineering estimation based on well choke pressure drop. One
representative well completion and one well work-over per field horizon must be developed to
characterize emissions per day of venting from all other completions and workovers in that field
horizon. This factor must be updated every two years. Dominion requests that the emission factor
be determined once for each horizon and updated at the operator's discretion based on the
operator's knowledge of that formation
Response: EPA has provided the least burdensome monitoring method that provides reasonable
accurate information. The field-specific emissions factors, however, have to be developed on a
once-in-two-years basis and reported on an annual basis. This is required for EPA to track
progress or deterioration in emissions levels. In addition, EPA cannot put verification and
auditing systems in place that allows for discretion of individual operations.
1395

(98.233 (a)(b)) High bleed devices must be calculated on a separate per device basis while low
bleed device calculations let you just put in a factor for all. Why not have a factor for high bleed
devices to make this part of the rule more uniform and easier to use
98.233(2),(h):
Gas well venting during unconventional well completions and workovers and gas well venting
during conventional well completions and workovers: EPA provides two calculation
methodologies for unconventional wells and one calculation methodology available for
conventional wells. For the purposes of calculating GHG emissions under Subpart W, there is no
valid reason to make a distinction between "conventional" and "unconventional" wells. We
request that EPA make available all three calculation methodologies for all well completions and
workovers, regardless of whether the well is considered "conventional" or "unconventional."
Response: EPA disagrees with the commenter. EPA has provided two different methodologies
for unconventional and conventional well completions and workovers (now referred to as well
completions from and without hydraulic fracturing) after careful consideration. Typically,
unconventional well completions and workovers require hydraulic fracturing that take longer and
vent more gas than do conventional well completions and workovers. Hence, EPA has provided
more detailed methods for well completions and workovers from hydraulic fracturing. Also, in
todays final rule, EPA has made the source categorization clearer by distinguishing between
well completions and workovers that do or do not require hydraulic fracturing. In addition, well
workovers that do not require fracturing use a simple emissions factor. With regards to the third
method in the TSD, please see the response to EPA-HQ-OAR-2009-0923-1011-11.
98.233(f) Well venting for liquids unloading
At 98.233(f)(1)(i), the calculation methodology for Well Venting for Liquids Unloading
requires the use of a recording flow meter to record the gas flows. ConocoPhillips requests EPA
add the option to use a well choke as in 98.233(g)(1)(ii). ConocoPhillips Alaska routinely uses
1396
the choke method to measure flow. This method is accurate and, more importantly, safe. When
gas vents, the devices used to measure that venting must be Class I, Division I approved. Even
such approval involves risk we prefer not to incur. Our practice has been to employ portable
skid-mounted pressure gauges or chokes that are not electric. We manually record the pressure
readings from these gauges or chokes and use that data to calculate the gas flow rates. We
request that EPA extend this ability to use non-electric gauges or chokes with manual recording
to all well work activities.
Response: EPA is not allowing a flow calculation across the well choke for liquids unloading.
See response to comment EPA-HQ-OAR-2009-0923-0956-2 for more information.
(98.233 (c)) Under natural gas driven pumps, it requires you to maintain a log of the amount of
liquid pumped. This brings up a larger question in terms of what type of accuracy is the EPA
looking for in these logs and how much supporting documentation do you need? (For example, if
you are using a diaphram pump to pump out storm water from diked areas, do we need to have a
meter on the pump or can we make an estimate based on rainfall that year
Response: In todays final rule, EPA has simplified the monitoring method for natural gas
driven pumps and required the use of emissions factors. For this monitoring method only a count
of the pneumatic pumps is required. For further details, please refer to Section II.F of the
(98.233 (f)(1)(C)(ii)) The equation assumes that the casing is vented to the atmosphere. Many
wells uses tubing strings with packers which limits the amount of blowdown gas. Should the
equation be modified to use vented tubular (pipe) diameter instead of casing diameter when
tubing and packing is in the well
Response: In todays final rule, EPAs method for calculating well venting for liquids unloading
does not assume that the casing is vented to the atmosphere directly. Rather, that liquid
accumulation in the tubing string, as it suppresses gas flow to the sales line, gradually shuts in
the well just the same as if the well were shut-in at the wellhead. Thus, reservoir flow slows and
pressure surrounding the well perforations builds to shut-in pressure, which fills the casing
outside the tubing with a bubble of gas at or near shut-in pressure. It is this bubble of high
pressure gas, in conjunction with removal of sales line back pressure, that provides a rush of gas
flow up the tubing during venting that attempts to lift the liquids accumulation to the surface.
EPA chose to not add complexity in the equations for well liquids unloading venting the effects
1397
of tubing packers. EPA understands tubing packers reduce the gas available to expel the liquids
and determined that the smaller bubble of gas inside the casing will compound the inefficiency
of lifting liquids without a plunger lift. EPA considers it likely that well venting for liquids
unloading may require continued venting beyond the average time that EPA calculated to expel
the full well depth of gas accumulation in the casing. Hence, to inform future policy in the most
cost-effective way, the simple equation including venting time would compensate for the
additional complexity of evaluating the gas bubble with myriad combinations of wells with
tubing packers. Please see EPA-HQ-OAR-2009-0923 Change to Rule Equation W-7: Time to
Vent the Casing Gas from Well Liquids Unloading for more details.

Commenter:
Source Type:
Well venting for liquids unloading
40 CFR 98.233(f)
Either Calculation Methodology 1 or 2 must be used to determine emissions for well venting.
The same method must be used for the entire reporting year.
Calculation Methodology 1:
1. Cumulative amount of time in hours of well venting during the year.
2. Flow Rate in cubic feet per hour, under ambient conditions as required in paragraph
(f)(1)(i)(A), (f)(1)(i)(B) and (f)(1)(i)(C) of this section.
Calculation Methodology 2:
1. Casing diameter (inches).
2. Well depth (feet).
3. Shut-in pressure (psig).
4. Number of vents per year.
5. Sales flow rate of gas well in cubic feet per hour.
6. Hours that the well was left open to the atmosphere during unloading.
Comments:
Meters are not designed to monitor intermittent flow. As liquids unloading has previously not
generally metered, it will be difficult to estimate the orifice size and/or meter capacity in order to
provide direct data to calculate this. As the process of installing meters for this purpose will be
experimental, Yates proposes that BAMM is used for this estimation for the first reporting year.
For smaller wells, calculation methodology 2 would be most appropriate, however the number of
hours the well was vented is difficult to monitor. Yates requests the use of BAMM for the first
1398
reporting year as operators attempt to resolve this issue.
Source Type:
Gas well venting during conventional well completions and workovers
40 CFR 98.233(h)
1. Daily gas production rate in cubic feet per minute.

2. Cumulative amount of time of well venting in minutes during the year.
Comments:
Yates will need to install flow meters for this requirement. Therefore Yates requests the use of
Source Type:
Gas well venting during unconventional well completions and workovers
40 CFR 98.233(g)
1. Daily gas production rate in cubic feet per minute.

2. Cumulative amount of time of well venting in minutes during the year.
Comments:
Yates will need to install flow meters for this requirement. Therefore Yates requests the use of
Response: In todays final rule, EPA provides options to use BAMM for certain sources for
specified time periods and allows reporters to petition for BAMM for other sources that meet
certain criteria. For further details, please see Section II.F of the preamble to todays final rule.

Commenter:
Comment on 98.233(f):
WBIH requests EPA reconsider the calculation methodology for well venting for liquids
unloading. The venting is a combination of gas and water and is not in a consistent proportion gas venting is variable
1399
Response: EPA agrees that the unloading is not a single phase flow. However, the monitoring
methods required provide a best possible approximation of emissions without unreasonable
burden for monitoring. Hence, EPA has retained the proposal monitoring method in todays final
rule.
Commenter:
Comment on 98.233(g):
WBIH requests EPA reconsider the calculation methodology for well venting for liquids
unloading. Gas venting is variable. For the well completions, the venting is a combination of gas,
water and sand; and for the workovers, air is injected during the process.
Response: EPA has provided adjustments to the monitoring methods to account for CO2 and
CH4 emissions only such that any air injected in a workover process does not get accounted as
emissions. For further clarification, please see the response to EPA-HQ-OAR-2009-0923-1074
32.
Commenter:
Comment on 98.233(h) and 98.233(l):
WBIH requests the removal of the word "venting" and the associated calculations for "venting".
Gas is not vented during testing - gas goes to the flare
Response: There are certain parts of the country where gas is not flared due to various reasons,
including safety and fire hazard. In such cases, gas from the testing is vented and hence EPA has
retained the term in todays final rule. Additionally, the use of the term or lack thereof does not
have any consequence on the monitoring method required.
Section 98.233(f) Well venting for liquids unloading.
Onshore Production Source Category: Well venting for liquids unloading
1400
API appreciates EPAs provision of alternate methodologies for estimating emissions from
venting of gas wells to unload liquids and urges EPA to retain both options. However, we believe
that the methodologies described by EPA in the proposed rule are unnecessarily complex,
burdensome and can be significantly improved while still yielding high quality data. As
structured in the proposed rule, operators are required to use one of two methods to determine
venting emissions. For clarity, each method and alternative is discussed separately.
Method #1 As Proposed by EPA: This method requires that operators measure emissions from
one vent per unique well tubing diameter and producing horizon/formation combination in each
gas producing field (presumed to be a named field as identified in the EIAs field list referenced
in the proposed rule) and then apply the venting rate to all other wells in the same unique
combination on a venting time basis. Due to the large number of onshore gas fields (estimated
as 15,715 covered fields), API estimates a total of 94,000 unique venting measurements would
be required to encompass the unique combinations of fields, producing horizons/formations,
tubing diameters, and operators. At a projected cost of $1,000 per measurement (a conservative
cost for temporary piping modifications, installation of a temporary recording flow-meter,
conducting the measurements, and returning the piping to its original configuration), the cost of
measurement would be $94 MM. Assuming 2.5 hours per field/tubing size/operator combination
(94,000) and $100 per hour to track venting by combination and perform the calculations for
emissions estimation, an additional cost of $23.5 MM is projected. API estimates a total cost of
$117.9 million to uniformly implement the Method #1 requirements, which would be fully
repeated every two years with a lower burden and cost in the interim years between
measurements. To address this problem, API suggests the following:
Method #1 Alternative as Proposed by API: Rather than requiring venting measurements for
each field, allow field groupings in a particular basin (Sub-basin entity) on a producing
horizon/formation specific basis with no distinction for different tubing sizes. In a producing
area, the production from a particular producing horizon/formation tends to be quite uniform and
operators tend to use the same or very similar tubing sizes. Also, the venting rate is much more
dependent on the producing horizon/formation pressure, reservoir flow dynamics, and operating
pressures (determined by the collection system pressures which tend to be uniform) than the
tubing size. Operators would establish a venting rate, based on the measurements for each Subbasin entity, and then track the total venting time on a producing horizon/formation basis to
arrive at an emissions estimate for the Sub-basin entity. For wells which are completed in and
producing from multiple producing horizons/formations, the venting rate would be assumed to
be the higher rate established for the applicable producing horizons/formations being produced.
Although the reduction in reporting burden is difficult to forecast, it could be substantial when
compared against the proposed EPA methodology.
Method #2 As Proposed by EPA: Method #2 requires calculation of the blow-down volume
from de-pressuring each individual well-bore and then adding the reservoir flow component
based on time and the individual wells production rate. This requires tracking, at an individual
well level, both the number and time of each individual venting event. With the large number of
gas wells subject to the rule (estimated as 373,000) and using EPAs estimate of the percentage
of these which vent (41.3%), API estimates about 154,000 individual well-bore de-pressuring
1401
calculations would be required. Using EPAs estimate of an average of 31 discrete vent instances
per venting well per year would require tracking of the number and time of some 4.7 million
discrete venting instances per year. To arrive at an emission estimate would require subsequently
calculating emissions for each of the 154,000 wells and then summing these emissions, on a
Basin entity level, for annual emissions. Assuming 0.5 hours total per well to calculate the wellbore de-pressuring volume for each affected well; 0.1 hours per event to track, QA/QC, and log
the time; and 0.2 hours per well to calculate and QA/QC the vent volumes to arrive at an
emissions estimate at $100 per hour, yields an estimated 293 man years of time (at 2,000
hours/yr) and $58.5 MM cost for this method.
The variables used in this method make it more difficult to apply to plunger lift venting and for
data gathering/assumptions. The issues with each of the variables are:
- Casing Diameter - The volume that would be vented during a plunger vent should only be the
volume of gas above the plunger in the tubing. The gas occupying the casing should have no
impact on the volume of gas that will be released during the vent. For plunger venting, using the
tubing diameter for the volumetric portion of the calculation would be more appropriate. This
method assumes the entire casing volume is vented each time, which is not the case for a normal
vent event on plunger lift wells.
- Well Depth - This variable is a decent input for the volume that would be released but as is
mentioned above, the volume above the plunger is all that should be vented. The plunger
normally runs to the bottom of the tubing where it impacts a spring set in the seating nipple. In
the worst case scenario, where a plunger is vented from the bottom of the tubing, the maximum
displacement is represented by the volume contained within the tubing which is dependent on the
length of that tubing. In some cases the depth of the tubing and well may not be significantly
different, but in others there could be a very large interval of open wellbore below the end of the
tubing. For this reason it would be more appropriate to use tubing depth rather that well depth
when determining the volume vented during a plunger vent.
- Shut-in Pressure Is this value the shut-in pressure right before the vent (casing or tubing?) or
the normal shut-in pressure of the reservoir? Both would require gathering additional vent or
reservoir data. Plunger wells are configured to vent only after a lifting period has elapsed. The
shut-in pressure of the well does not accurately represent the pressure of the tubing directly
before a vent occurrence. In order to determine the volume of gas that will be vented from the
tubing it is necessary to know the pressure of the gas occupying the tubing, this would be more
accurately represented by tubing pressure or line pressure if tubing pressure is not available.
- Sales Flow Rate We find no logic or justification for including the sales volume in addition to
the vented volume. Provided the plunger lift controller equipment is counting vent events
properly, the sales flow rate should not be a consideration in determining the plunger vent
volumes. Plunger vents should only be venting the gas above the plunger which is represented by
the volumetric portion of the calculation. Applying the sales rate would artificially increase the
reported volume. This would also require gathering flow rate data for each well at the time of the
vent event.
1402
As prescribed, Method 2 would require the collection of more vent data (pressure, rate and vent
time) and overestimate the vent volume due to the use of the entire casing volume plus sales flow
rate. If the method were altered to use entire tubing volume, tubing pressure and omitted the
sales flow rate it would be more applicable to plunger lift wells.
Method #2 Alternative as Proposed by API: Rather than requiring the calculation of the wellbore de-pressuring volume for each individual well which vents in a grouping of fields (Sub
basin entity) within a Basin entity, require the calculation of the well-bore de-pressuring volume
for a typical well-bore configuration and pressure for each unique producing horizon/formation.
This calculated volume would be applied to all wells producing from the unique producing
horizon/formation in the Sub-basin entity. The reservoir flow component would be additive
rather than on an individual well production rate basis, based on the total venting time of all
wells producing from the unique producing horizon/formation and the average production rate of
all wells producing from the unique producing horizon/formation. For wells which are completed
in and producing from multiple producing horizons/formations, both the well-bore de-pressuring
volume and the reservoir flow per unit of time would be assumed to be the higher rate
established for the applicable producing horizons/formations being produced.
Response: EPA agrees with the commenter that use of a plunger lift changes the calculation of
emissions for method #2. Todays final rule includes a third method for venting a well to the
atmosphere with the aid of a plunger lift. In this calculation, the volume of gas in the tubing
string above the plunger, at sales line pressure, is calculated for an average half-hour period to
bring the plunger and liquid load to the surface. If the well is left open longer than this average
half-hour, then the continued venting to the atmosphere is estimated at average sales flow rate.
This is a cost-effective alternate to measuring this flow rate. EPA recognizes that the gas bubble
accumulated in the well casing at shut-in pressure (which will occur gradually as liquids
accumulation in the tubing string gradually stop gas flow to the sales line) will flow up the
unimpeded tubing string to the atmosphere at a higher rate than normal flow to the sales line.
However, this is very difficult to calculate costly to measure. Therefore, to adequately inform
future policy, the average sales flow rate is something known for each well, and a conservative
estimate for those cases where wells are left open longer than necessary to raise the plunger. For
more information on this additional well liquids unloading method, please see EPA-HQ-OAR
2009-0923 Change to Rule Equation W-7: Time to Vent the Casing Gas from Well Liquids
Unloading.
Also, please see the comment EPA-HQ-OAR-2009-0923-1151-96 for a response to the
commenters cost estimates for monitoring well liquids unloading.

Section 98.233(g) Gas well venting during unconventional well completions and workovers
API appreciates EPAs provision of alternate methodologies for estimating emissions from flow
1403
back and clean-up of wells following fracture stimulation during initial completion and
subsequent work-over and urges EPA to retain this flexibility. However, we believe that the
methodologies described by EPA in the proposed rule are unnecessarily complex, burdensome
and can be significantly improved while still yielding high quality data. As structured in the
proposed rule, operators are required to use one of two methods to determine completion and
work-over venting or flaring emissions. For clarity, each method and alternative is discussed
separately.
Method #1 As Proposed by EPA: This method requires that operators measure the flow rate from
one completion event and one work-over event in each gas producing field (presumed to be a
named field as identified in the EIAs field list referenced in the proposed rule) and then apply
the flow rate to all other completions and work-overs on a flow-back time basis. Due to the
large number of onshore gas fields (estimated as 15,715 covered fields) API estimates a total of
7,072 completion event measurements and 3,929 work-over event measurements would be
required. This would be followed by tracking flow-back time for each of some 39,000
completion and work-over events and then calculating emissions. API estimates a total cost of
$15.0 million to uniformly implement the Method #1 requirements which would be fully
repeated every two years with a lower burden and cost in the interim years between
measurements.
Method #2 As Proposed by EPA: This method requires that operators estimate the flow rate from
named field as identified in the EIAs field list referenced in the proposed rule) by measuring the
pressure drop across a choke and then performing a choke-flow calculation to arrive at an
average flow-rate. This rate would then be applied to all other completions and work-overs on a
flow-back time basis. Due to the large number of onshore gas fields (estimated as 15,715
covered fields) API estimates a total of 7,072 completion event estimates and 3,929 work-over
event estimates would be required. This would be followed by tracking flow-back time for each
of some 39,000 completion and work-over events and then calculating emissions. API estimates
a total cost of $9.5 million to uniformly implement the Method #2 requirements which would be
fully repeated every two years with a lower burden and cost in the interim years between
measurements.
To address the problems outlined for EPAs proposed Methods 1 and 2, API suggests the
following alternative:
Alternative Method Proposed by API: Rather than requiring completion and work-over estimates
for each field, require them for field groupings in a particular basin (Sub-basin entity) on a
producing horizon/formation specific basis. In a producing area the reservoir characteristics and
behavior from a particular producing horizon/formation tends to be quite uniform and operators
tend to use the same or very similar hydraulic fracture and well clean-up techniques and
practices. Although the reduction in burden is difficult to forecast it could be substantial when
compared against the proposed EPA methodology.
API also requests that EPA clarify what type of events are within the scope of coverage for this
category, particularly for work-overs. It is not clear what well work events classify as work
1404
overs subject to the requirements of the rule. Section 98.233 (g) Gas well venting during
unconventional well completions and work-overs appears to limit the calculation of emissions
to well clean-up post hydraulic fracturing operations with the following language: Calculate
emissions from gas unconventional well venting during well completions and work-overs from
hydraulic fracturing. Other down-hole well work, such as tubing changes, maintenance, etc, do
not appear to be covered. API agrees with EPA that the proper focus is on the high rate flowback following fracture stimulation but requests that this be made clearer in the rule.
The rule also needs to make provision for incorporating the flow back of energizing gas used in
energized fracs. The two typical gases used in energized fracs are nitrogen and carbon dioxide.
When a well is flowed for clean-up following an energized frac, the energizing gas will be the
majority of the gas component flowed back during the initial stages of clean-up. Most/all of the
energizing gas is recovered from the formation during clean-up and is flared or vented. This gas
is purchased by weight and the quantity is known. API recommends that EPA include a
provision in Section 98.233 (g) which attributes the appropriate portion of the flow from a well
clean-up following an energized frac to the energizing gas used. In the case of nitrogen it would
be subtracted from the total measured or calculated flow. In the case of carbon dioxide, the
portion equal to the amount of gas pumped would be subtracted from the total flow and
evaluated as pure carbon dioxide rather than the typical well stream gas mixture.
Response: EPA disagrees with the commenter. With regard to methodologies for estimating
emissions from gas well venting during well completions and workovers with hydraulic
fracturing, please see the response to EPA-HQ-OAR-2009-0923-1305-46.
Also, please see the comment EPA-HQ-OAR-2009-0923-1151-97 for a response to the
commenters cost estimates for monitoring well venting from completions and workovers with
hydraulic fracturing.
Section 98.233(h) Gas well venting during conventional well completions.
Onshore Production Source Category: Gas well venting during conventional well completions
Section 98.233 (h) requires calculating emissions from each gas well venting during
conventional well completions and workovers based on the daily gas production rate and
cumulative amount of time each well is venting during the year.
For new wells, which have not established a production rate, the rule should allow the use of
the average flow rate from the first 30 days of production.
Response: EPA agrees with this comment and is now allowing in todays final rule the use of
the average flow rate from the first 30 days of production for new wells. In the event that the new
1405
well is completed less than 30 days from the end of the reporting period, the first 30 days of
production straddling the current and following reporting years shall be used.
Section 98.233(l) Well testing venting and flaring
Onshore Production Source Category: Well testing venting and flaring
Section 98.233(l) requires determining the gas to oil ratio (GOR) of the hydrocarbon production
for each well tested, determining the flow rate of oil for the well being tested, and tracking the
number of days in the year that the well is tested.
The vast majority of well tests are conducted while the wells are in operation and do not require
venting or flaring. As a result, the rule should exempt well testing that does not result in venting
or flaring natural gas.
Response: In response to the commenters concerns on well testing and flaring of natural gas, the
source monitoring method description in the supplementary proposed rule sufficiently clarifies
that the emissions are to be reported only when there is venting or flaring.
Section 98.233(f)(1)(ii) Well venting for liquids unloading . API suggests that it would be
appropriate to include the use of API Compendium Equation 5-28 as an alternative to Equation
W-7.
E [subscript CH4 or CO2] = (9.78 x 10^-7) x (1 / Z) x (D [subscript casing])^2 (P) x (CH4 or CO2
mole fraction) x (MW [subscript CH4 or CO2]) x (# blowdowns / year) x (tones / 2204.62 lb)
Response: EPA disagrees that the commenter proposed equation adequately accounts for well
venting for liquids unloading. This equation is basically the same concept as that proposed in
todays final rule for well venting without the aid of a plunger lift, with the exception of the
venting time element. EPA has determined from Gas STAR partner feedback that liquids
unloading without the aid of a plunger lift is very inefficient, increasingly so with lower shut-in
pressure. Because of this inefficiency, it is possible that well operators may leave the vent valve
open for a longer time that that necessary to just exhaust the gas bubble in the well casing at
shut-in pressure. Therefore, in todays final rule, in addition to the option of metering the gas
vent, EPA is providing two equations for liquid unloading venting to the atmosphere, each
1406
including a time element: one equation without a plunger lift and one with the aid of a plunger
lift.

Well Venting
Due to the dramatic under-reporting of well venting emissions, as described in Section 2 of our
comment above, and the profitable opportunity to collect emissions at a sizable profit (e.g.
Green Completions or Reduced Emissions Completions (REC), we recommend that EPA
require Operators to meter vented gas volume at each well.
We recognize that there are some cases where it may not be technically feasible to implement a
green completion. For example, well pressure may be too low or pipeline infrastructure may not
yet be installed (e.g. exploratory drilling or the first well on a well site). Gas pipelines are
typically not installed until it is confirmed that an economic gas supply is found. Therefore, gas
from the first well is often flared or vented during drilling and completion activities because
there is not a pipeline in place. However, subsequent wells drilled on that same pad would be in a
position to implement green completion techniques. In these limited cases, we would support
EPAs proposed well venting emission estimating methods that are formation specific (field
horizon) and based on tubing size, flow rate and venting duration, with three important
modifications. We agree with WCIs recommendation that the frequency of the sampling and
metering should be increased to obtain representative data for wells where reservoir or well
conditions are unique.
EPA proposes that one representative well completion and one representative well workover per
field horizon be developed to characterize daily venting emissions from all other completions
and workovers within a given field horizon. A single well measurement, however, may not be
representative of the entire field and therefore a more statistically robust sample size (e.g., 3-5
wells, or 5% of wells, etc.) should be adopted. EPA proposes that this field level well venting
emission factor be updated once every two years.
We agree with WCIs recommendation that the frequency of the sampling and metering should
be increased to obtain representative data for wells where reservoir or well conditions are unique.
We recommend that the emission factor be updated annually, because well flowrates can vary
substantially over time.
For unconventional wells (e.g., hydraulically fractured wells), if the fracturing is typically done
in multiple stages and is in fact not a continuous process, then measuring the emissions in units
of days of venting may not be as good as requiring measurement in units of hours of
venting. EPA should consider requiring reporting in the most representative units of time.
1407
Response: EPA notes the commenters position in regard to reduced emissions completions and
has retained the option to use a meter for estimating emissions. However, a vast majority of the
wells are still not completed using reduced emissions completions and hence EPA has retained
the option to use a choke calculation to estimate emissions.
EPA notes the commenters concern on accuracy of monitoring methods and the need for
stringent quality requirements but EPA does not agree with the commenter. With regards to
accuracy of monitoring methods, please see the response to EPA-HQ-OAR-2009-0923-0582-25.
With regards to the comment on the units of time, EPA does not see this as an issue. Hour units
can be converted to day units by dividing the total hours by 24. EPA has used units that are most
convenient to most of the reporters.
Gas well venting during unconventional well completions and workovers: Noble supports the
API comment regarding gas well venting during unconventional well completions and
workovers. Additionally, Noble requests, as proposed by API, the flexibility of allowing the
completion and work-over measurements (Method #1) or estimates (Method #2) to be completed
for an even larger grouping. This grouping could be multiple producing horizons/formations,
rather than a single producing horizon/formation, if the reservoir characteristics and behavior
from the group of horizons/formations tend to be quite uniform and operators tend to use the
same or very similar hydraulic fracture and well clean-up techniques and practices.
Response: EPA disagrees with the commenter. With regards to well completion and workover
groupings, please see the response to EPA-HQ-OAR-2009-0923-1151-19.
Gas well completions and well workovers.
As previously noted, emissions from well completions (based on industry experience as reported
through EPAs Natural Gas STAR program) are probably significantly underestimated 194 in
each unique field. In fact, the emissions factor being used in the 2008 U.S. GHG Inventory is
194
See also 75 FR 18621: [N]o body of data has been identified that can be summarized into generally applicable
emissions factors to characterize emissions from these sources [(i.e., from well completion venting and well
workover venting).
1408
believed to significantly underestimate emissions based on industry experience as included in the

EPA Natural Gas STAR Program publicly available information (http://www.epa.gov/gasstar/).
In addition, the 2008 U.S. GHG Inventory emissions factor was developed prior to the boom in
unconventional well drilling (1992) and in the absence of any field data and does not capture the
diversity of well completion and workover operations or the variance in emissions that can be
expected from different hydrocarbon reservoirs in the country. ] The 2008 U.S. GHG Inventory
emphatically states that [n]atural gas well venting due to unconventional well completions and
workovers, as well as conventional gas well blowdowns to unload liquids have already been
identified as sources for which Natural Gas STAR reported reductions are significantly larger
than the estimated inventory emissions. 195 And EPA has again indicated, in response to
questions from Environmental Defense Fund, that [p]resently, [reduced emission completions
(REC)] reductions reported in the Natural Gas STAR body of work is larger than well
completion venting in the inventory on an annual basis. 196 Specifically, the U.S. GHG
Inventory is based on an emission factor of a little over three thousand standard cubic feet (3
Mcf) per gas well drilled and completed. 197
Yet Natural Gas STAR program partner experience shows several cases where emission factors
were thousands of times higher than those shown in the 2008 inventory. Examples include (1) a
BP project employing green completions at 106 wells and reporting 3,300 Mcf of gas recovered
per well 198 (2) a Devon Barnett Shale project employing green completions at 1,798 wells
between 2005 and 2008 and reporting 6,300 Mcf of gas recovery per well, 199 and (3) a Williams
project employing green completions at 1,064 wells in the Piceance Basin reporting 23,000 Mcf
of gas recovered per well. 200
All of these examples include gas recovery estimates more than 1,000 times higher than the 3
Mcf of gas per well estimated in the U.S. GHG Inventory for 2008.
These data are consistent with the unconventional gas well completion and workover data
presented in EPAs TSD. Specifically, Appendix B on pp. 79-82 includes four examples from
the Natural Gas STAR program with gas completion rates of 6,000 Mcf, 10,000 Mcf, 700 Mcf
195
Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006, p.
3-47.
196
See Environmental Defense Fund (EDF) Questions & USEPA Answers June 1, 2010 posted to the Docket
for this action on June 2, 1010, Docket ID EPA-HQ-OAR-2009-0923-0070.
197
Table A- 118: 2008 Data and CH4 Emissions (Mg) for the Natural Gas Production Stage, p. A-144, Inventory of
U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006.
198
See Natural Gas STAR Program Recommended Technologies and Practices for Wells at
http://www.epa.gov/gasstar/documents/green_c.pdf, slide 11.
199
See Attachment B, 2009 workshop presentation by Devon, slides 3 and 13. 6,300 Mcf = 11.4 Bcf / 1,798 wells.
200
See Natural Gas STAR Program Recommended Technologies and Practices for Wells at
http://www.epa.gov/gasstar/documents/workshops/pennstate2009/robinson1.pdf, slide 16. NOTE: the reported
Williams reductions appear to assume an average of 32 days of uncontrolled venting during flowback, which
seems like an unreasonably long length of time.
1409
and 20,000 Mcf per completion. EPA used an average of these four data points for its emission
factor for this source (i.e., 9,175 Mcf/completion). Using this factor resulted in estimated
emissions from completions of conventional and unconventional wells of 86 billion standard
cubic feet (Bcf). By comparison, the U.S. GHG Inventory reports emissions of just 0.1 Bcf of
gas from drilling and well completions. 201.202
More generally, Natural Gas STAR partners reported recovering between 7 and 12,500 Mcf
(average of 3,000 Mcf) of natural gas from each cleanup with the potential for an estimated 25
Bcf of natural gas recovery from green completions annually in 2005, compared with the annual
U.S. GHG Inventory total for Drilling and Well Completion of just 0.1 Bcf in 2008. And more
recently, EPAs Natural Gas STAR program attributed 45 Bcf of gas to reduced emissions
completions (RECs) in 2008 (representing 50 percent of EPAs Natural Gas STAR programs
annual total reductions).203 If, in fact, the emission factor for well completions is at least 1,000
times higher than what is reported in the U.S. GHG Inventory (e.g., if it is at least 3,000 Mcf
instead of 3 Mcf), this adds 100 Bcf to the total estimated emissions from natural gas systems
raising the total from 240 Bcf to 340 Bcf in 2008 (a 40% increase). 204
A New York Times article published on October 15, 2009, reports that EPA is currently
reviewing and revising methane emissions from U.S. gas wells 205
According to the article:
An E.P.A. review of methane emissions from gas wells in the United States strongly implies that
all of these figures may be too low. In its analysis, the E.P.A. concluded that the amount emitted
by routine operations at gas wells not including leaks like those seen near Franklin is 12
times the agencys longtime estimate of nine billion cubic feet. In heat-trapping potential, that
new estimate equals the carbon dioxide emitted annually by eight million cars. 206
In fact, the TSD for the MRR includes an estimate of 120 Bcf for U.S. well completion and
201
See http://www.epa.gov/gasstar/documents/green_c.pdf, slides 9 and 5; Table A-125: CH4 Emission Estimates

from the Natural Gas Production Stage Excluding Reductions from the Natural Gas STAR Program and NESHAP
regulations (Gg), p. A-151, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S.
EPA # 430-R-10-006. Drilling and Well Completion (2008) = 2.05 Gg.
202
2.05 Gg. (2.05 Gg * 0.052 ft3/g = 0.1 Bcf).
203
See 2009 EPA Natural Gas STAR Program Accomplishments, available online at
http://www.epa.gov/gasstar/documents/ngstar_accomplishments_2009.pdf. Total sector reductions (2008) = 89.3
Bcf of which 50% are the result of RECs (50% of 89.3 Bcf = 45 Bcf).
204
See Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2008 (April 2010) U.S. EPA # 430-R-10-006,
Chapter 3, Table 3-38 CH4 Emissions from Natural Gas Systems (Gg), p. 3-45. Total emissions in Bcf: 4,591 Gg
* 0.052 ft3/g = 240 Bcf.
205
206
1410
workover venting 207 As previously discussed, this estimate is based on an emission factor of
9,175 Mcf per completion or workover. Given the sparse data (4 data points) and the enormously
wide range of potential emission factor values from this source (ranging from 700 to 20,000
Mcf), even the recently revised estimates may very well continue to underestimate emissions
from this source.
The statistical representation of the U.S. GHG Inventory data includes a 95 percent certainty
range within which emissions from this source category are likely to fall for the year 2008. This
range, for Natural Gas Systems, includes a lower bound of -24% and an upper bound of
+43%. 208 As noted in the uncertainty analysis, [t]he heterogeneous nature of the natural gas
industry makes it difficult to sample facilities that are completely representative of the entire
industry. 209 Basing an emission factor on only a few representative sources with highly
variable rates results in a potentially high degree of uncertainty, as reflected in the reported
uncertainty range.
The EPA-acknowledged discrepancies between the inventory emissions reported to-date and the
Natural Gas STAR-reported reductions (and the high degree of uncertainty in both of these data
sources) demonstrate the need for more reliability in emissions reporting from the oil and gas
sector.
The Large and Growing Number of Oil and Gas Sources Means Continuous Improvement is
Necessary to Minimize Uncertainty in the Reported Data.
Certain oil and gas sources are of particular importance due to the magnitude of their emissions.
These sources will require the best possible reporting requirements if the data gathered are to be
representative and of further use. In particular, uncertainties in data reported across a large
population of sources (e.g., the hundreds of thousands of pneumatic devices used throughout the
oil and gas sector) have the potential to greatly reduce the value of the dataset. Pneumatic
devices are a prime example of a source that is particularly important in terms of magnitude and
is also important due to the sheer number of individual sources, each one with uncertain levels of
emissions. The magnitude and importance of this emissions source support a greater emphasis on
direct measurement (e.g., metering) and on rigorous verification and audit for any reporting that
will be based on engineering estimates (e.g., OEM emission factors).
Response: EPA agrees with the commenter that certain sources, such as well venting, are
underestimated in the national inventory. EPA has where possible balanced between cost to
report and the need for better quality data. In the case of well liquids unloading, well
completions, and well workovers EPA has retained the monitoring methods proposed. However,
for pneumatic devices, considering the costs to monitor high bleed devices and pneumatic
207
See TSD at 82.
208
Chapter 3, Table 3-41, p. 3-46.
209
Chapter 3, p. 3-46.
1411
pumps, EPA has required the use of emissions factors in todays final rule. See major changes in
the rule in Section II.F of the preamble to todays final rule for more details.
Gas well venting during unconventional well completions and workovers & Gas well venting
during conventional well completions and workovers
Section 98.233(g) & (h): EPA provides two calculation methodologies for unconventional wells
and one calculation methodology available for conventional wells. For the purposes of
calculating GHG emissions under Subpart W, there is no valid reason to make a distinction
between conventional and unconventional wells. IPAMS requests that EPA make available
all three calculation methodologies for all well completions and workovers, regardless of
whether the well is considered conventional or unconventional.
Response: EPA disagrees with the commenter. With regards to making all three calculation
methodologies available to conventional and unconventional wells, please see the response to
EPA-HQ-OAR-2009-0923-1040-20.
Section 98.233(g) Gas well venting during unconventional well completions and workovers and
(h) Gas well venting during conventional well completions and workovers
For the well completion emissions, the rule does not seem to specifically address green
completions and how these are to be handled. The rule should be clarified to only include flow
which is emitted to atmosphere or flared.
Response: EPA agrees with the commenter. In todays final rule, EPA allows for adjustments
for completion and workover hydrocarbons being sent to a flare.
Section 98.233(f) Well venting for liquids unloading.
BP appreciates EPAs provision of alternate methodologies for estimating emissions from
venting of gas wells to unload liquids and urges EPA to retain both options. However, we believe
1412
that the methodologies described by EPA in the proposed rule are unnecessarily complex,
burdensome and can be significantly improved while still yielding high quality data. As
structured in the proposed rule, operators are required to use one of two methods to determine
venting emissions. For clarity, each method and alternative is discussed separately.
Method #1 As Proposed by EPA: This method requires that operators measure emissions from
one vent per unique well tubing diameter and producing horizon/formation combination in each
gas producing field (presumed to be a named field as identified in the EIAs field list referenced
in the proposed rule) and then apply the venting rate to all other wells in the same unique
combination on a venting time basis. Due to the large number of onshore gas fields (estimated
as 15,715 covered fields), a large number of unique venting measurements would be required to
encompass the unique combinations of fields, producing horizons/formations, and tubing
diameters. At a projected cost of $1,000 per measurement, for temporary piping modifications,
installation of a temporary recording flow-meter, conducting the measurements, and returning
the piping to its original configuration, the burden and cost of measurement would be significant.
To address this problem, BP suggests the following:
Method #1 Alternative as Proposed by BP: Rather than requiring venting measurements for each
field, require them for field groupings in a particular basin (Sub-basin entity) on a producing
horizon/formation specific basis with no distinction for different tubing sizes. In a producing
area, the production from a particular producing horizon/formation tends to be quite uniform and
operators tend to use the same or very similar tubing sizes. Also, the venting rate is much more
dependent on the producing horizon/formation pressure and reservoir flow dynamics along with
the flowing operating pressures (determined by the collection system pressures which tend to be
uniform) than the tubing size. Operators would establish a venting rate, based on the
measurements, for each sub-basin entity and then track the total venting time on a producing
horizon/formation basis to arrive at an emissions estimate for sub-basin entity. For wells which
are completed in and producing from multiple producing horizons/formations, the venting rate
would be assumed to be the higher rate established for the applicable producing
horizons/formations being produced. Although the reduction in burden is difficult to forecast it
would be substantial when compared against the proposed methodology.
Method #2 As Proposed by EPA: Method #2 requires calculation of the blow-down volume
from de-pressuring each individual well-bore and then adding the reservoir flow component
based on time and the individual wells production rate. This requires tracking, at an individual
well level, the number and time of each individual venting event. With the large number of gas
wells subject to the rule a very large number of individual well-bore de-pressuring calculations
would be required. This would be followed by tracking of a much larger number of discrete
venting instances per year in order to calculate emissions.
Method #2 Alternative as Proposed by BP: Rather than requiring the calculation of the wellbore
de-pressuring volume for each individual well which vents, in a grouping of fields (Reporting
Unit/Area) within a basin require the calculation of the well-bore de-pressuring volume for a
typical well-bore configuration and pressure for each unique producing horizon/formation. This
calculated volume would be applied to all wells producing from the unique producing
horizon/formation in the Reporting Unit/Area. The reservoir flow component would be added
1413
based on the total venting time of all wells producing from the unique producing
horizon/formation and the average production rate of all wells producing from the unique
producing horizon/formation rather than on an individual well production rate basis. For wells
which are completed in and producing from multiple producing horizons/formations, both the
well-bore de-pressuring volume and the reservoir flow per unit of time would be assumed to be
the higher rate established for the applicable producing horizons/formations being produced.
E. Section 98.233(h) Gas well venting during conventional well completions.
Section 98.233 (h) requires calculating emissions from each gas well venting during
conventional well completions and workovers based on the daily gas production rate and
cumulative amount of time each well is venting during the year.
For new wells, which have not established a production rate, the rule should allow the use of
the average flow rate from the first 30 days of production.
F. Section 98.233(g) Gas well venting during unconventional well completions and workovers
BP appreciates EPAs provision of alternate methodologies for estimating emissions from
flowback and clean-up of wells following fracture stimulation during initial completion and
subsequent work-over and urges EPA to retain this flexibility. However, we believe that the
methodologies described by EPA in the proposed rule are unnecessarily complex, burdensome
and can be significantly improved while still yielding high quality data. As structured in the
proposed rule, operators are required to use one of two methods to determine completion and
work-over venting or flaring emissions. For clarity, each method and alternative is discussed
separately.
BP also requests that EPA clarify what type of events are within the scope of coverage for this
category, particularly for work-overs. It is not clear what well work events classify as
workovers subject to the requirements of the rule. Section 98.233 (g) Gas well venting during
unconventional well completions and workovers appears to limit the calculation of emissions to
well clean-up post hydraulic fracturing operations with the following language: Calculate
emissions from gas unconventional well venting during well completions and workovers from
hydraulic fracturing. Other down-hole well work, such as tubing changes, maintenance, etc, do
not appear to be covered. BP agrees with EPA that the proper focus is on the high rate flow-back
following fracture stimulation but requests that this be made clearer in the rule.
The rule also needs to make provision for incorporating the flow back of energizing gas used in
energized fracs to assist with post frac clean-up. The two typical gasses used in energized fracs
are nitrogen and carbon dioxide. When a well is flowed for clean-up following an energized frac,
the energizing gas will be the majority of the gas component flowed back during the initial stages
of clean-up. Most/all of the energizing gas is recovered from the formation during clean-up and
is vented to atmosphere (either while flaring or directly venting). This gas is purchased by weight
and the quantity is known. BP recommends that EPA include a provision in 98.233 (g) which
attributes the appropriate portion of the measured or calculated flow from a well clean-up
following an energized frac to the energizing gas used. In the case of nitrogen it would be
subtracted from the total measured or calculated flow. In the case of carbon dioxide, the portion
1414
equal to the amount of gas pumped would be subtracted from the total measured or calculated
flow and evaluated as pure carbon dioxide rather than the typical well stream gas mixture.
Method #1 As Proposed by EPA: This method requires that operators measure the flow rate from
named field as identified in the EIAs field list referenced in the proposed rule) and then apply
the flow rate to all other completions and work-overs on a flow-back time basis. Due to the
large number of onshore gas fields (estimated as 15,715 covered fields) this would require a
large number of measurements be made initially and then repeated every two years. To address
this problem, BP suggests the following:
Method #1 Alternative as Proposed by BP: Rather than requiring completion and work-over
measurements for each field require them for field groupings in a particular basin (Reporting
Units/Areas) on a producing horizon/formation specific basis. In a producing area the reservoir
characteristics and behavior from a particular producing horizon/formation tends to be quite
uniform and operators tend to use the same or very similar hydraulic fracture and well clean-up
techniques and practices. Although the reduction in burden is difficult to forecast it would be
substantial when compared against the proposed methodology.
Method #2 As Proposed by EPA: This method requires that operators estimate the flow rate from
named field as identified in the EIAs field list referenced in the proposed rule) by measuring the
pressure drop across a choke and then performing a choke-flow calculation to arrive at an
average flow-rate. This rate would then be applied to all other completions and work-overs on a
flow-back time basis. Due to the large number of onshore gas fields (estimated as 15,715
covered fields) this would require a large number of measurements be made initially and then
repeated every two years. To address this problem, BP suggests the following:
Method #2 Alternative as Proposed by API: Rather than requiring completion and work-over
estimates for each field require them for field groupings in a particular basin (Reporting
Units/Areas) on a producing horizon/formation specific basis. In a producing area the reservoir
characteristics and behavior from a particular producing horizon/formation tends to be quite
uniform and operators tend to use the same or very similar hydraulic fracture and well clean-up
techniques and practices. Although the reduction in burden is difficult to forecast it would be
substantial when compared against the proposed methodology.
Response: In developing the petroleum and natural gas onshore production segment, EPA
considered many alternative facility definitions, ranging from potential groupings of wells
connected to gathering stations, to fields, to basins, all with one principle in mind: provide a
definitive definition that all producers can directly apply without arbitrary or arcane discretion.
Fields are defined and catalogued, as are basins. Any sub-basin grouping of fields by various
combinations of similarities in production quality, oil gravity, gas to oil ratio, shut-in pressure,
formation permeability and porosity, age and location would be significantly less than definitive,
and therefore, EPA retained the basin definition for onshore production facilities.
The option of methods #1 and #2 provided by EPA are both practical and cost-effective. Method
#2 is more cost effective, and requires simply well and field data that is readily available and
1415
easy to analyze to determine well liquids unloading venting, provided the time of venting is
collected and retained. Where well venting is performed with a fixed timer plunger lift, the
venting time is a matter of noting the timer setting and manual changes in timer setting. For
advanced programmable logic controller optimized cycle venting, the PLC can be programmed
to retain the number and duration of venting events. For manual venting to the atmosphere, it is
incumbent on the operator to log the event and time between opening and closing the vent to
atmosphere. As noted by this commenter, in EPA provided method #2 , the total venting time of
all wells producing from a unique producing horizon/formation is a part of the equation for
extended venting at average well production rate. Therefore, this data contemplated by the
commenter as a component of the venting calculation, only on a field basis in todays final rule
rather than an undefined combination of fields. Therefore, in developing the petroleum and
natural gas onshore production segment EPA decided against any reporter developed
combinations of fields within a basin as a basis for emissions calculation methods. EPA also
disagrees with the commenter that the reporting from well liquids unloading sources is
burdensome; please see response to EPA-HQ-OAR-2009-0923-1151-96 for further details.
EPA agrees and is now allowing in todays final rule the use of the average flow rate from the
first 30 days of production for new wells. For further information, see response to comment
EPA-HQ-OAR-2009-0923-1151-120.
For changes to the methodology for well completions and workovers in todays final rule please
see the response to comment EPA-HQ-OAR-2009-0923-1011-11. EPA has determined that the
monitoring of well completions and workovers with hydraulic fracturing per the supplementary
proposed rule methods is not significantly burdensome. Please see response to EPA-HQ-OAR
2009-0923-1151-97 for details on how the number of samples required for monitoring well
completions and workovers with hydraulic fracturing is manageable for all reporters. Since the
burden to report is manageable and grouping of fields is not required, EPA has retained the
monitoring methods for well completion and workovers.
In todays final rule, EPA has revised the source type name for well workovers based on whether
or not there is a fracturing involved. Well workovers that do not require fracturing use a simple
emissions factor.
EPA agrees that energized fracs are emerging, and affect the quantity of GHG emissions
(methane and CO2, depending on whether they are CO2 gas assist or nitrogen gas assist). Todays
final rule includes an adjustment of the vented GHG emissions for gas composition as
determined by sample analysis, available gas composition data or, in the specific case of gas well
completions with energized hydraulic fracturing, a material balance based on accounting for the
energized gas fraction of the total gas emissions.
1416
13.2.2
TANK EMISSIONS

10. De minimis thresholds should be added to the modeling requirements in
98.233(j) and the direct measurement requirements in 98.233(k), and subjections (j) and
(k) should be revised. Section 98.233(j) would require emissions modeling from production and
processing storage tanks, while 98.233(k) would require direct measurement of emissions from
transmission storage tanks. There is no reason to subject de minimis emission sources to the
onerous and costly reporting requirements that would be required by 98.233(j) and (k). The
rule should be modified so that the requirements of subparagraph (j) and (k) would not apply
unless the tank at issue met a 5 bbl/day threshold for throughput of hydrocarbon condensate. 210
In addition, owners or operators of components that are required to meet the requirements
of subparagraph (j) should be able to use HySYS or Promax as an alternative to E&P Tank:.
These technologies are at least as sophisticated as E&P Tank: and are widely recognized and
accepted in the industry. As such they should be available as alternatives. Similarly, the list of
specific data parameters in subparagraph (j)(I) should be modified to allow operators to select the
appropriate representative data for whichever authorized software package they employ.
Proposed 98.233(k)(2) would require direct measurement if a stuck dump valve was
detected at a transmission storage tank. This requirement should be eliminated, even if other
direct measurement requirements are retained. A stuck dump valve is a malfunction that, when
discovered, will quickly be repaired. It is akin to a leak along a pipeline, which as EPA states
will quickly be repaired and, as a result, is not subjected to Subpart W at all. There is no reason
to impose a metering requirement on a malfunction that when discovered will be fixed promptly.
Indeed, the ironic result of the rule in its current state would be that such immediate repairs could
not be made - because the leak would have to be measured before it was fixed. This makes no
sense and is contrary to the purpose of the environmental laws in general, which is to reduce
emissions rather than exacerbate them.
Response: EPA agrees that onshore production storage tanks with a throughput below a certain
equipment threshold should be given relief from using complex process models and simulations
to determine GHG emissions. Hence, EPA has set an equipment threshold at 10 barrels per day
that would result in approximately 18% of tanks to report emissions using a software program
and 82% of tanks to report emissions using an emission factor. EPAs selected equipment
threshold is more conservative than the commenters recommended 5 barrel per day equipment
threshold. Please see the memo Equipment Threshold for Tanks in the rulemaking docket
210
TPA's preference is that any direct measurement requirements in the rule (such as in 98.233(k) be eliminated
altogether.
1417
(EPA-HQ-OAR-2009-0923) for how EPA determined this threshold. , EPA determined that
setting an equipment threshold of 10 barrels per day of separator throughput, such that any tank
connected to a separator below the equipment threshold use a simple emissions factor approach,
would result in significant reduction in burden without significantly compromising emissions
data reporting quality.
EPA reviewed the comment related to setting an equipment threshold on transmission storage
tanks and disagree that a threshold is necessary. There are relatively fewer storage tanks in the
transmission segment; as low as one or two tanks per compressor station. In addition, operators
or owners of transmission storage tanks will only have to conduct direct measurement and report
emissions if continuous emissions are detected from the vapor vent stacks using an optical gas
imaging instrument. EPA through its Natural Gas STAR Program has determined that
malfunctioning dump valves can be a source of significant emissions through liquid tanks.
Hence, given the low burden and importance of characterizing this source of emissions, EPA has
retained the requirements for transmission storage tanks. The information gathered through this
requirement will help EPA characterize the frequency of such occurrences in the industry to
suitably inform policy .
EPAs intent in requiring the use of E&P Tank in the rule was to standardize the calculation
methods across reporters. However, EPA notes the prevalence of other process simulators and
modeling tools in use in the industry and understands that there is some equivalency in their
modeling capabilities. Hence, in todays final rule, EPA does not prescribe E&P Tank or any
other specific process simulator. Instead, EPA allows the use of any process simulators that
meets the requirements under Calculating GHG emissions Onshore production and processing
storage tanks.
EPA disagrees with the commenter that monitoring of stuck dump valves should be removed.
EPA agrees that stuck dump valves are usually repaired upon detection and the rule does not
preclude operators from fixing the malfunctioning valves. Specifically, if a scrubber dump valve
is found malfunctioning during standard operations at a time other than the official leak survey,
nothing in the rule states that the reporter must not fix the dump valve or take a measurement.
The only time when a measurement must be taken is if a scrubber dump valve is found leaking
during an official leak survey. However, prior to detection, these stuck dump valves are
emitting gas, which could be significant source of emissions for some facilities. EPA recognizes
that the dump valve will remain open during the monitoring but this is needed to accurately
estimate the emissions. In fact, EPA contends that such malfunction will not be widespread in
facilities where the current maintenance practices result in minimal occurrences of such
malfunctions, and should be no cause for concern. Finally, EPA has provided the option of using
acoustic leak detection and measurement devices that can quickly detect and measure leaks
without the need for more burdensome and time consuming direct measurement approach. This
should mitigate any concerns on the emissions from dump valves due to the monitoring
requirements set forth in todays final rule.
1418

Process Simulators: In some cases the rule allows the use of simulation software packages;
however, because the exact software package is specified [such as GRI GLYCalc in Section
98.233(e) and API E&P TANK in Section 98.233(j)], suitable alternatives are excluded. There
are other commercially available, industry-accepted process simulators that are as accurate as (or
more so) than the prescribed software packages (e.g., HYSYS, ProMax, PRO/II, PROSIM, etc.).
In addition, these process simulators can be used for source categories that do not currently have
simulation software packages recommended in the rule (e.g., AGR vents). IPAMS requests that
EPA not specify the simulation software package for each source category, and instead allow all
commercially available, industry-accepted process simulators.
Response: EPA is allowing the use of alternate process simulators to estimate emissions from
onshore production storage tanks. Please see the response to EPA-HQ-OAR-2009-0923-1061-10
for further details. EPA agrees with the commenter on the use of process simulation software
program for AGR vent. Please see the response to EPA-HQ-OAR-2009-0923-1024-26 for
further details.
The proposed rule only allows for the use of GRI GLYCalc and E&P Tanks. Other commonly
used modeling programs include HYSYS, Vasquez Beggs Equations, and process modeling
software. Please consider allowing data from these more sophisticated programs.
onshore tanks and dehydrators. Please see the response to EPA-HQ-OAR-2009-0923-1061-10.
Commenter:

Additional engineering estimation methods should be included.
EPA has specifically identified two process simulation models in the proposed Subpart W, GRIGLYCalc for estimating dehydrator emissions and E&P Tanks for estimating production and
processing tank emissions. Many companies use more sophisticated process simulation models
for some facilities to estimate emissions from tanks and acid gas removal units (AGR unit). The
two most common are AspenTechs HYSYS process simulator and Bryan Research and
Engineerings ProMax process simulator. These process simulators are widely accepted as the
gold standards for detailed process design and optimization in the gas processing, chemical and
refining industries.
1419
onshore production. Please see the response to EPA-HQ-OAR-2009-0923-1061-10 for further
details.
EPA is allowing the use of alternate process simulators to estimate emissions from dehydrator
vent; please see the response to EPA-HQ-OAR-2009-0923-0053-4.
Commenter:

Minimum thresholds should be applied for equipment level applicability and calculations.
The EPA does not incorporate thresholds for applicability or for using simplified calculations at
the equipment level. As a result, in many cases complex process models and simulations are
required for equipment with minimal GHG emissions.
The first example is related to tanks, for which the same annual calculation method is prescribed
regardless of the tanks throughput, for example a tank that fills at a rate of 1 barrel/day versus a
tank that fills at a rate of 100 barrels/day. In this particular example, the prescribed method
requires analysis of the separator liquid and collection of numerous operating parameters. The
liquid analysis is expensive and often the sample is difficult or impossible to collect, even with
higher liquid production rates, and should not be required for tanks with minimal throughput.
Across the natural gas industry, there are thousands of tanks associated with very low liquid
production rates and, therefore, minimal GHG emissions. GPA recommends that tanks with a
throughput of less than 5 barrels/day be excluded from reporting under Subpart W. If an
exclusion cannot be provided, GPA recommends tanks below the 5 barrel/day threshold be
allowed to use the appropriate emission factor from the API Compendium of Greenhouse Gas
Emission Methodologies for the Oil and Gas Industry (API Compendium) in place of conducting
modeling using process simulations.
Response: EPA agrees that determining input parameters that require sampling of low-pressure
separator oil is difficult and burdensome. In todays final rule, EPA has provided alternate
methods to determine parameters that require direct measurement of the low-pressure separator
oil, namely the composition and Reid vapor pressure. These parameters can be determined using
default values provided in some process simulators, previous analyses that are representative of
the reporting facility, or direct measurement.
EPA agrees with the commenters recommendation to include an equipment threshold for
onshore production storage tanks and use of an emissions factor. Please see the response to
EPA-HQ-OAR-2009-0923-1061-10.
1420

The proposed rule does not contain emissions calculations for tanks that primarily contain water
except for transmission station condensate tanks. The Preamble (pg. 18,621) requests comments
on how to quantify emissions from transmissions tanks storing water without creating additional
reporting burden. Xcel Energy requests that transmission station condensate tanks be excluded
from reporting due to the de minimis nature of these emissions in the system. Quantification of
emissions from transmission condensate tanks would be burdensome to reporters; the cost of
required resources would far outweigh the benefit of this information.
Response: EPA disagrees that transmission station condensate tanks constitute a small emissions
source and therefore should be eliminated from the rule. EPA recognizes that emissions from
transmission storage tank condensate itself as a result of vapors flashing from the condensate
and/or water is minimal. Rather, GHG emissions resulting from the gas blowing through a
malfunctioning scrubber dump valve through a storage tank can be large and therefore it is
EPAs intent to collect information on emissions from this source. As a result, operators or
owners of transmission storage tanks will only have to conduct direct measurement if continuous
emissions are detected from the vapor vent stacks using an optical gas imaging instrument or
acoustic leak detection device; see response to EPA-HQ-OAR-2009-0923-1061-10 for further
details. The monitoring methodology in todays final rule does not require the calculation of
flashing losses, only emissions from leaking dump valves.
Commenter:
EPA seeks comments on how to quantify emissions from tanks storing water without resulting in
additional reporting burden to the facilities. (page 66)
MidAmerican submits that requiring the direct measurement of emissions from water
storage tanks would be overly burdensome. There are proven analytical methods that are
capable of calculating greenhouse gases released in storage tanks. However, it is crucial to
differentiate between storage tanks for natural gas production and natural gas transmission.
Response: In regards to emissions from water tanks please see response to EPA-HQ-OAR-2009
1421
Water Storage Tanks

EPA seeks comments on how to quantify emissions from tanks storing water without resulting in
additional reporting burden to the facilities.
BP supports EPAs stated goal of quantifying emissions without resulting in additional reporting
burden and believes that EPA should not include quantification of emissions from tanks storing
water in Subpart W. If EPA does choose to include this very small source, the default emission
factors provided in the API Compendium for production and transmission condensate should be
sufficient for providing a simple emission estimation method for this small source.
Response: EPA agrees with the commenter and has not required the estimation of emissions
from water storage tanks in onshore production in todays final rule. However, tanks that store
water/condensate in gas transmission segment are still required to report emissions; see response

Tank Emissions
We support WCIs recommendation that production and processing storage tank emissions be
estimated using the gas-oil-ratio (GOR). The GOR measurement should be taken at a location
immediately prior to the storage tank.
We strongly support EPAs proposed emissions reporting method for transmission storage tanks
that uses an optical gas imaging device and gas metering.
Response: With regard to the WCIs recommendation, please see the response to EPA-HQ
OAR-2009-0923-0582-11.
13.2.2.1
ONSHORE PRODUCTION AND PROCESSING TANKS

Section 98.233(j) Storage tanks
BP supports EPAs selection of modeling as the methodology for determining emissions from
Storage Tanks and supports the use of E&P Tanks as an applicable model. However, BP has
1422
several comments to clarify this category in the rule along with several proposals to broaden the
methodologies available and to streamline the determination of GHG emissions from Storage
Tanks thus significantly reducing the burden imposed by the proposed rule.
1. To clarify the scope of this source category, EPA should modify 98.233 (j) to apply only
to storage tanks receiving produced hydrocarbon liquids rather than produced liquids as
currently worded. All of the discussion in the preamble, the TSD, the calculation
methodology in 98.233 (j), and the functionality of E&P Tanks apply only to
hydrocarbon liquids. This clarification would make it clear that this source category does
not encompass produced water tanks.
2. As proposed, the rule requires sampling and modeling of emissions from each individual
separator dumping hydrocarbon liquids to a tank at atmospheric pressure. This presents a
very large sampling, analysis, and modeling burden. Below, in these comments, BP
presents alternatives to address the dilemma of applying the quantification requirements
to each individual separator/tank combination.
3. As proposed, the rule restricts the selection of methods to modeling using E&P Tanks
latest software version. BP believes the allowable methodologies should be broadened as
follows:
a. Allow the use of E&P Tanks, Version 2 or higher as noted in 98.7.
b. Allow the use of process simulation models, such as HySys, ProSim, etc. as
alternatives to using E&P Tanks. These models are in wide use within the
industry, are the standard for process design and optimization, and have much
broader capabilities than E&P tanks.
c. Allow the use of a laboratory GOR determination rather than modeling of
separator tank combinations.
4. BP requests that EPA eliminate the adjustment multiplier provisions applied to the
output of E&P Tanks as specified in 98.233 (j) (4). Based on the discussion of the
derivation of these multipliers in the TSD, it appears that EPA is attempting to quantify
abnormal equipment malfunctions of dump valves sticking and routing raw gas to the
storage tank along with presumed vortexing of raw gas to the tank while dumping. The
adjustment multipliers were apparently derived from examining a ratio of the output of
E&P Tanks runs, using the GEO-RVP method, to measured emissions from the recent
TCEQ tank study report. BP observes that the correlations of the same E&P Tank runs to
the TCEQ data have very low correlation constants (0.0719 and 0.0045 for crude oil and
condensate respectively) in the TSD discussion of how they were derived. Although no
evaluation of the accuracy of the adjustment multipliers is presented in the TSD, it
seems logical that they would share the same low correlation between the two methods
and therefore have no basis. To compound this issue, as EPA is aware, there is substantial
doubt, even within TCEQ, that the TCEQ study was conducted correctly and the data is
suspect. As EPA observed in the TSD, equipment malfunction and vortexing have not yet
been thoroughly studied or quantified, the sample data set is limited (and of dubious
accuracy), and only very weak correlations were observed for the available data. BP also
notes that the evaluation presented in the TSD was done using E&P Tanks in the GEO
RVP mode while the rule specifies running the simulation based on actual pressurized
separator hydrocarbon liquid samples and conditions which further degrades any already
dubious application of these adjustment multipliers.
1423
In addition to the very weak information and understanding that underpins this
requirement, as currently proposed the provision in 98.233 (j) (4) directs operators to
apply the adjustment multipliers in the case where the tank emissions are not
represented by the equilibrium conditions of the liquid in a gas-liquid separator and
calculated by E&P Tank. It is completely unclear to how an operator would identify
such a case and hence implement the adjustment multipliers.
Based on all of these readily apparent reasons, BP does not believe that the justification
or adequate knowledge to apply the provisions of 98.233 (j) (4) exist and these should be
removed from the rule.
To address the dilemma of having to sample and model each individual separator and
tank combination and the resulting burden and cost, BP suggests two alternative
approaches which will significantly reduce this burden while still yielding high quality
data. In a particular subbasin entity within a hydrocarbon basin, the hydrocarbon liquid
production from the same or very similar producing horizons/formations is very similar
in composition, properties, and behavior and does not vary widely from one well to
another. As EPA is aware, for a given producing well/site and fluid compositions, the
separator pressure and differential pressure between the separator and the hydrocarbon
liquid tank is the largest determinant of emissions from the tank, including CO2 and
Methane. This characteristic allows a stratified sampling of separator/tank combinations
within a Reporting Unit/Area to be used to develop a robust methodology of quantifying
emissions from individual sites without sampling and modeling each separator/tank
combination. Descriptions of each of these approaches follow:
Proposed Option 1: Sampling Without Modeling
Based on a significant number of hydrocarbon liquid sample flashing behavior using HYSYS,
we have determined that greater than 95% of the CO2 in pressurized hydrocarbon liquids and
greater than 98% of the methane in hydrocarbon liquids is released when the material is
dumped to an atmospheric pressure storage tank. BP recommends that EPA not require
modeling of tanks for emissions of methane and CO2, but simply assume that all methane and
CO2 in pressurized condensate/crude is released to the atmosphere when the material is flashed
to ambient pressure. This will eliminate the need for modeling of each tank as specified in the
rule.
To address the case where tank emissions are routed to a combustion device (flare, incinerator,
etc) for control, specify development of a multiplier, developed from a small number of tank
modeling simulations in a sub-basin area, to be applied to the methane portion CO2 yield when
combusted to estimate combustion of the total stream.
To address the burden posed by sampling (at pressure) and analyzing pressurized liquid from
each individual separator in the industry, BP suggests that EPA enable a stratified sampling
within a particular sub-basin entity as follows:
1. Determine the operating pressure range of the separators which dump hydrocarbon
liquids to atmospheric tanks within the Reporting Unit/Area.
1424
2. Select the lowest and highest pressure separators along with three additional sites with
separator pressures which are approximately equally distributed between the lowest and
highest pressure.
3. Obtain pressurized samples and analyses for each of the selected separators.
4. Directly calculate CO2 and CH4 emissions from the pressurized crude/condensate based
on the total.
5. Plot the emissions versus separator pressure separately for Methane and Carbon Dioxide
(if there are emissions) in Excel using the Scatter Plot option.
6. Click on the plotted points and select Add Trendline from the menu.
7. On the Type tab, select the trendline type which gives the reasonable best fit curve as
defined by the highest R-squared value (avoid very high order polynomial curves). To
show the R-squared value and resultant equation, On the Options tab, select Display
equation on chart and Display R-squared value on chart, then select OK.
8. Use the produced equation to calculate the emissions for each separator/tank combination
based on the site annual throughput and separator pressure. Company records, such as
sales gas pressure measured at the site sales meter, shall be used to determine separator
pressures.
9. Determine if vapor recovery units or combustion control devices are used for each tank in
the population. If a combustion device is used determine the average destruction
efficiency and apply it to the equation output and determine resultant emissions using the
requirements of 98.233 (n). If a VRU is used, apply the estimated gas recovery
percentage to the equation output to adjust the emissions determined by the equation.
Option 2: Regression Curve Storage Tank Approach for a Reporting Unit/Area
1. Determine the operating pressure range of the separators which dump hydrocarbon
liquids to atmospheric tanks within the Reporting Unit/Area.
2. Select the lowest and highest pressure separators along with three additional sites with
separator pressures which are approximately equally distributed between the lowest and
highest pressure.
3. Obtain pressurized samples and analyses for each of the selected separators along with all
data required by 98.233 (j) (1) (i) through (viii) if using E&P tanks as the modeling suite
for emissions estimation. If using an alternative methodology (providing EPA enables
options) collect all data necessary to use these alternative methodologies.
4. Run the modeling suite or laboratory GOR and determine emissions of methane and CO2
in pounds per sales oil (liquid) barrel.
5. Plot the emissions versus separator pressure separately for Methane and Carbon Dioxide
(if there are emissions) in Excel using the Scatter Plot option.
7. On the Type tab, select the trendline type which gives the reasonable best fit curve as
defined by the highest R-squared value (avoid very high order polynomial curves). To
show the R-squared value and resultant equation, On the Options tab, select Display
equation on chart and Display R-squared value on chart, then select OK.
8. Use the produced equation to calculate the emissions for each separator/tank combination
based on the site annual throughput and separator pressure. Company records, such as
sales gas pressure measured at the site sales meter, shall be used to determine separator
pressures.
1425
9. Determine if vapor recovery units or combustion control devices are used for each tank in
the population. If a combustion device is used determine the average destruction
efficiency and apply it to the equation output and determine resultant emissions using the
requirements of 98.233 (n). If a VRU is used, apply the estimated gas recovery
percentage to the equation output to adjust the emissions determined by the equation.
Exclusion or Simplified Estimation for Small Tanks
BP recommends that storage tanks at individual well sites or central tank batteries with a
throughput of less than 5 barrels/day be excluded from reporting under Subpart W. If an
exclusion cannot be provided, BP recommends storage tanks at individual well sites or central
tank batteries with a throughput below the 5 barrel/day threshold be allowed to use the
appropriate emission factor from the API Compendium of Greenhouse Gas Emission
Methodologies for the Oil and Gas Industry (API Compendium) in place of conducting sampling
and analysis and/or modeling using process simulations.
Response: EPA agrees with the commenter on monitoring only hydrocarbon liquid tanks and
has clarified text in todays final rule to indicate that only emissions from hydrocarbon liquid
flashing is required to be monitored under the onshore production storage tanks source type.
Flashing emissions from dissolved hydrocarbons in produced water is minimal and the burden
associated with reporting from such a small source of emissions is not justified.
With regard to the use of alternate process simulators to estimate emissions from onshore
production storage tanks, EPA has provided options on the use of software programs; see the
EPA has reviewed the commenters suggestion to eliminate the use of an adjustment multiplier
and determined that the multiplier is a more cost-effective approach for informing future policy
than requiring direct measurement. There is no simple way to simulate or calculate how much
gas could bypass in a stuck-open dump valve or be entrained by vortex in the liquid drain from
the separator to the storage tank. EPAs information from Natural Gas STAR partners and
studies by the Houston Advanced Research Center and Texas Counsel on Environmental Quality
indicate that sticking, leaking or vortexing through dump valves is not an uncommon cause of
excessive gas emissions from field tanks. The fact that these emissions may go un-noticed or
undetected for long periods in remote, un-manned operations suggests that basing tank emissions
solely on separator liquid properties could significantly under-estimate those emissions. EPAs
goal with inclusion of the adjustment multiplier is to inform policy in two regards: activity data
on how common this cause of excessive emissions is occurring, and adjustment of liquid flashing
emissions for this bulk gas bypass. EPA agrees that the conclusions in the TCEQ study are
incomplete. However, the methodology and procedures used to measure the emissions from the
storage tanks are acceptable making the data suitable for EPAs analysis. The HARC and TCEQ
studies are the best-available, public data on crude oil stock tanks which revealed that
approximately 50% of the tanks tested exhibited this phenomenon. Other data publicly reported
in Natural Gas STAR workshops indicated that stuck-open scrubber dump valves were in one
case the largest methane emission source in transmission compressor stations, and contributed
nearly 90% of emissions in a transmission compressor station site survey (Natural Gas STAR
Compressors Technology Transfer Webcast, January 18, 2006). Natural Gas STAR partners have
1426
responded to questions in workshops, stating that dump valve malfunction resulting in gas
bypass is a problem that does occur.
In todays final rule, EPA does not require reporters to arbitrarily determine whether conditions
in the tanks are actually at equilibrium conditions. Instead, in todays final rule, reporters are
required to report occurrences of dump valves not closing during the calendar year. Because the
gas that would bypass the separator level control valve is an important revenue stream for the
operation, EPA concluded that production operators will be attentive to sales volumes of oil and
gas, and recognize when separator dump valves are malfunctioning. Industry experts readily
agree that this problem exists, more so in some fields.
EPA disagrees that the low correlation coefficient observed in the first method evaluated by the
EPA to characterize emissions from tanks not at equilibrium conditions, renders the second
method invalid. The adjustment multipliers in todays final rule are not related to the low
correlation coefficients observed in the first method. In fact, the EPA justifies its usage of the
second methodology over the first methodology based on these low correlation coefficients.
EPA has considered the methodologies suggested by the commenter and has decided to include
the sampling without modeling option in todays final rule. EPA conducted a similar analysis
using E&P Tank and observed that over 95% of the methane and carbon dioxide is emitted. As a
result, EPA is allowing operators to determine the composition of the low-pressure separator oil
using a standard method published by a consensus-based standards organization and assume that
all of the CH4 and CO2 are emitted. This method requires no modeling and where such oil
composition at separator pressure and temperature exists, this alternative significantly reduces
the burden while maintaining the same data quality as the modeling methodology. While EPA
agrees with the sampling without modeling option, EPA does not agree with the comment on
aggregation of fields within a basin level reporting. For further details, please see the response to
EPA-HQ-OAR-2009-0923-1305-46.
The EPA chose not to include the commenters regression curve in todays final rule. For
clarification on the issues regarding regression curves, please see the response to EPA-HQ-OAR
2009-0923-1305-52.
onshore production and processing storage tanks. For further clarification, please see the

Section 98.233(j) Storage tanks.
API has several comments to clarify this category in the rule along with several proposals to
1427
broaden the methodologies available and to streamline the determination of GHG emissions from
storage tanks thus significantly reducing the burden imposed by the proposed rule.
i. Source Category Scope
To clarify the scope of this source category, EPA should modify Section 98.233 (j) to
apply only to storage tanks receiving produced hydrocarbon liquids, rather than
produced liquids as currently worded. All of the discussion in the preamble, the TSD,
the calculation methodology in Section 98.233 (j), and the functionality of E&P
Tanks apply only to hydrocarbon liquids. This clarification would make it clear that
this source category does not encompass produced water tanks.
ii. Modeling Methodology
API supports EPAs selection of modeling as the methodology for determining
emissions from Storage Tanks and supports the use of E&P Tanks as an applicable
model. However, as proposed, the rule restricts the selection of methods to modeling
using E&P Tanks latest software version. API believes the allowable methodologies
should be broadened as follows:
- Allow the use of E&P Tanks, Version 2 or higher as noted in Section 98.7.
- Allow the use of process simulation models or industry accepted estimating
methods as alternatives to using E&P Tanks. These models and methods are in
wide use within the industry, are the standard for process design and
optimization, and have much broader capabilities than E&P Tanks.
- Allow the use of a laboratory GOR determination rather than modeling of
separator tank combinations.
iii. Proposed Adjustment Multiplier
API requests that EPA eliminate the adjustment multiplier provisions applied to the
output of E&P Tanks as specified in Section 98.233 (j) (4). Based on the discussion
of the derivation of these multipliers in the TSD, it appears that EPA is attempting to
quantify abnormal equipment malfunctions of dump valves sticking and routing
raw gas to the storage tank along with presumed vortexing of raw gas to the tank
while dumping. The adjustment multipliers were apparently derived from
examining a ratio of the output of runs from E&P Tanks, using the GEO-RVP
method, to measured emissions from the recent TCEQ tank study report. API
observes that the correlations of the same E&P Tank runs to the TCEQ data have very
low correlation constants (0.0719 and 0.0045 for crude oil and condensate,
respectively) in the TSD discussion of how they were derived. Although no
evaluation of the accuracy of the adjustment multipliers is presented in the TSD, it
would seem that they share the same low correlation between the two methods, and
therefore have no basis. As EPA may be aware, to compound this issue, there is
substantial doubt, even within TCEQ, that the TCEQ study was conducted correctly
and the data is suspect. As EPA observed in the TSD, equipment malfunction and
vortexing have not yet been thoroughly studied or quantified, the sample data set is
limited (and of dubious accuracy), and only very weak correlations were observed for
the available data. API also notes that the evaluation presented in the TSD was done
using E&P Tanks in the GEO-RVP mode while the rule specifies running the
simulation based on actual pressurized separator hydrocarbon liquid samples and
conditions which further degrades any already dubious application of these
1428
adjustment multipliers.
In addition to the very weak information and understanding that underpins this
requirement, as currently proposed, the provision in Section 98.233 (j) (4) directs
operators to apply the adjustment multipliers in the case where the tank emissions
are not represented by the equilibrium conditions of the liquid in a gas-liquid
separator and calculated by E&P Tank. It is completely unclear to API and its
members how an operator would identify such a case and hence implement the
adjustment multipliers.
Based on all of these readily apparent reasons, API does not believe that EPA has any
justification or sufficient background data to apply the provisions of Section 98.233
(j) (4) and therefore these provisions should be removed from the rule. If EPA is
interested in pursuing this type of event occurrence, API could participate in a limited
cooperative study.
iv. Exclusion or Simplified Estimation for Small Tanks
For tanks, the same annual calculation method is prescribed regardless of the tank
throughput, for example a tank that fills at a rate of 1 barrel/day versus a tank that fills
at a rate of 100 barrels/day. API recommends that storage tanks at individual well
sites or central tank batteries with a throughput of less than 5 barrels/day be excluded
from reporting under Subpart W. If an exclusion cannot be provided, API
recommends storage tanks at individual well sites or central tank batteries with a
throughput below the 5 barrel/day threshold be allowed to use the appropriate
emission factor from the API Compendium in place of conducting modeling using
process simulations.
v. APIs Proposed Emission Estimation Alternatives
As proposed, the rule requires sampling and modeling of emissions from each
individual separator transferring hydrocarbon liquids to a tank at atmospheric
pressure. API estimates that this will require sampling of the pressurized separator
hydrocarbon liquid at some 307,734 individual separator/tank combinations, followed
by modeling of each individual tank. As currently configured, E&P tanks (or other
process simulation models) cannot be used in a batch processing mode and each
individual separator/tank combination would have to be individually set-up and
modeled. API estimates the burden of the required sampling ($92.3 MM) and
modeling ($15.4 MM) to be some $107.7 million for the covered portion of the
industry.
To address the dilemma of having to sample and model each individual separator and
tank combination with the attendant burden and cost, API suggests two alternative
approaches which will significantly reduce this burden while still yielding high
quality data. In a particular Sub-basin entity, the hydrocarbon liquid production from
the same or similar producing horizons/formations is very similar in composition,
properties, and behavior, and does not vary widely from one well to another. For a
given producing well/site and fluid compositions, the separator pressure and
1429
differential pressure between the separator and the hydrocarbon liquid tank is the
largest determinant of emissions from the tank, including CO2 and CH4. This
characteristic allows a stratified sampling of separator/tank combinations within a
Sub-basin entity to be used to develop a robust methodology for quantifying
emissions from individual sites without sampling and modeling each separator/tank
combination. Descriptions of each of these approaches follow.
Method #1 Alternative as Proposed by API: Sampling Without Modeling
Based on modeling the flashing behavior of a significant number of hydrocarbon liquid
samples using a process simulation model, one API member has determined that greater
than 95% of the CO2 in pressurized hydrocarbon liquids and greater than 98% of the
methane in hydrocarbon liquids is released when the material is dumped to an
atmospheric pressure storage tank. API recommends that EPA not require modeling of
tanks for emissions of methane and CO2, but simply assume that all methane and CO2 in
pressurized condensate/crude is released when the material is flashed to ambient pressure.
This will eliminate the need for modeling each tank as specified in the rule.
To address the burden posed by sampling (at pressure) and analyzing pressurized liquid
from each individual separator in the industry, API suggests that EPA enable a stratified
sampling within a particular Sub-basin entity as follows:
1. Determine the operating pressure range of the separators which dump
hydrocarbon liquids to atmospheric tanks within the Sub-basin entity.
2. Select the lowest and highest pressure separators along with three additional sites
with separator pressures which are approximately equally distributed between the
lowest and highest pressure.
3. Obtain pressurized samples and analyses for each of the selected separators.
4. Directly calculate CO2 and CH4 emissions from the pressurized crude/condensate
based on the total.
5. Plot the emissions versus separator pressure separately for CH4 and CO2 (if there
are emissions) to determine the R-squared value.
6. Use the produced equation to calculate the emissions for each separator/tank
combination based on the annual throughput and separator pressure. Company
records, such as sales gas pressure measured at the sales meter, shall be used to
determine separator pressures.
7. Adjust the emissions as necessary to account for the use of combustion control
devices or vapor recovery.
Method #2 Alternative as Proposed by API: Regression Curve Storage Tank Approach for a Subbasin Entity
APIs second proposed estimation method follows the same approach as Option 1, with
the exception of Steps 3 and 4 as outlined below. With this approach, modeling or
emission estimation methods are applied to the stratified sampling of separators across
the Sub-basin entity to develop a correlation equation relating emissions to volume of
sales oil.
3. Obtain pressurized samples and analyses for each of the selected separators along
with all data required by Section 98.233 (j) (1) (i) through (viii) for using E&P
1430
tanks as the emission estimation model. If using an alternative process simulation

model or industry accepted estimation methodology, collect all data necessary to
use these alternative methodologies.
4. Run the modeling suite or laboratory GOR and determine emissions of CH4 and
CO2 in pounds per sales oil (liquid) barrel.
API has developed examples of these optional estimation methods and could provide this
information to EPA separately.
Response: EPA has clarified the text under onshore production storage tanks that reporting is
only required from tanks storing hydrocarbon liquids. For further details, please see the response
to EPA-HQ-OAR-2009-0923-1305-48.
In todays final rule, EPA is allowing the use of alternate process simulators to estimate
emissions from onshore production storage tanks under certain conditions. For further
clarification, please see the response to EPA-HQ-OAR-2009-0923-1061-10.
In todays final rule, EPA disagrees with the commenters suggestion to eliminate the use of an
adjustment multiplier. For further clarification, please refer to the response to EPA-HQ-OAR
2009-0923-1305-48.
In todays final rule, EPA is allowing the sampling without modeling option suggested by API.
For further clarification, please refer to the response to EPA-HQ-OAR-2009-0923-1305-48.
In todays final rule, EPA has not included the regression curve option to estimate emissions. For
further details, please refer to the response to EPA-HQ-OAR-2009-0923-1305-52.
In todays final rule, EPA has included an equipment threshold for onshore production storage
tanks. For further details, please see the response to EPA-HQ-OAR-2009-0923-1061-10.

WCI Modifications: WCI recommends the method in SECTION 98.233(j) for calculating
emissions from onshore production and processing storage tanks is replaced by the following
method:
WCI recommends that the results of a periodic laboratory determination of GOR (Gas Oil Ratio)
from a pressurized liquid sample (2009 API Compendium, #2 on p. 5-41) be used to quantify
methane emissions at storage tank batteries where daily production exceeds a threshold value
(>10 bbl/day).
A pressurized liquid sample will be collected annually at a point downstream of all field
separators and prior to the point where produced liquid is flashed to atmospheric pressure (that
is, where it enters the storage tank). In the laboratory, the sample is allowed to reach atmospheric
pressure, the volume of gas generated measured (i.e. GOR determined) and a sample of the
evolved gas collected and analyzed for methane and carbon dioxide content. Additional testing is
required when a well is connected to or disconnected from the battery, or changes in separator
1431
operational parameters are made. Emissions calculated in this manner assume that the methane
generated in this measurement is ultimately released to the atmosphere (where there is no vapor
recovery in-place). At storage tank batteries where production is below the threshold (<10
bbl/day) E&P TANKS would be used to determine flow and vapor composition.
The assumption implicit with this approach is that only flashing losses are important for methane
emissions from storage tanks. While small amounts of methane and carbon dioxide will be
emitted during subsequent storage and handling, all the methane emissions quantified by this
method will be attributed to the initial depressurized storage tank. Storage tanks equipped with
vapor recovery units (VRU) are exempt, and measurements are limited to land-based storage
tanks containing condensate and crude oil.
Methane and CO2 emissions at storage tank batteries where the oil production rate is 10 barrels
per day or greater shall be calculated in the following manner:
E CH4/CO2 = GOR * PR * 1/MVC * MWg * MF CH4/CO2* 0.001
Where:
E CH4/CO2
GOR
PR
MVC
MWg
MF CH4/CO2
0.001
=
=
=
=
=
=
=
methane or carbon dioxide emissions (metric tonnes/year)

Gas Oil Ratio (scf/bbl)
oil production rate (bbl/measurement period)
molar volume conversion
molecular weight of the gas (kg/kg-mole)
mass fraction of methane or carbon dioxide in gas (kg CH4/kg gas)
conversion factor (metric tonnes/kg)
Methane and carbon dioxide emissions at storage tank batteries where the oil production rate is
less than 10 barrels per day shall calculate methane emissions using the E&P TANKS Model.
Sampling Protocol:
Sample collection for GOR determination.
A pressurized sample of produced liquids shall be collected from the separator at a location
upstream of the storage tank. This point would typically be at the final separation device before
produced oil transitions from separator outlet pressure to atmospheric pressure and enters a
production storage tank. This may require the installation of a sampling valve at the appropriate
location.
Sampling protocol specific to the collection of separator liquid can be found in the following
publications:
1. Appendix C Sampling Protocol section (pg 33) of the E&P TANK Version 2.0 Users
Manual. API is considering revisions to this software as the current version is out of
date with current EPA TANKS software.
2. Wyoming Department of Environmental Quality Air Quality Division guidance
document entitled Oil and Gas Production Facilities, Chapter 6, Section 2 Permitting
1432
Guidance (revised August 2001), Appendix D Sampling and Analysis of

Hydrocarbon Liquids and Natural Gas.
http://deq.state.wy.us/aqd/Oil%20and%20Gas/GUIDANCE2001.pdf
3. Gas Processors Association (GSA) Standard 2174-93, available for purchase at:
www.gasprocessors.com/product.asp?dept_id=2500&sku=S+2174-93
The sample collection pressure shall be determined at the time of collection and again prior to
processing in the laboratory to insure that sample integrity has been maintained. Liquid
temperature should also be determined and recorded at the time of collection.
Sampling and laboratory based determination of GOR shall be conducted at prescribed intervals
and at a time when operational parameters of the storage tank battery are representative and
consistent with normal operating conditions.
[Table] Sampling interval (n in Eq. 1)
An additional sample shall be collected and analyzed if:

1. the oil production rate at the storage tank battery changes more than 20% for time
periods in excess of one week (e.g., in cases where a well or wells feeding the storage
tank battery stop or start production).
2. separator operating pressures change by more than 10%
Determination of volume of liquid produced
The volume (barrels) of liquid produced during the sampling interval shall be determined using a
calibrated liquid meter or industry standard method to an accuracy of +/-5%.
Response: EPA agrees with the commenters suggested methodology and has included it as an
option. Please see the response to EPA-HQ-OAR-2009-0923-1305-48 for further details.

Section 98.233(j) Onshore production & processing storage tanks
El Paso estimates that gathering the data necessary to comply with the requirements for onshore
storage tanks is one of the most costly sources relative to the emissions contribution. EPA has
not considered this important aspect of the Subpart W rules as proposed and should consider
alternative approaches to these sources. The proposed rule requires the following parameters for
onshore production storage tanks:
1433
Separator oil composition;
Separator temperature;
Separator pressure;
Sales oil API gravity;
Sales oil production rate;
Sales oil reid vapor pressure;
Ambient air temperature; and
Ambient air pressure.
Current sampling practices are not detailed enough to support modeling input requirements.
Industry standard practice only collects basic compositional information and production
measurements for liquids, while E&P Tanks requires full speciation data from an extended
laboratory analysis. This liquid analysis is expensive and often the sample is difficult to collect
since it requires gathering a sample at the separator pressure. For some separators, the hardware
is not in place to collect a pressurized sample. In addition, specific safety precautions must be
employed to gather the liquid sample.
-
It is El Pasos interpretation that engineering estimates can be used for the required operational
parameters, since the rule does not specify metering and monitoring for these parameters. El
Paso requests confirmation of this interpretation. In addition, El Paso requests that tanks with a
throughput of less than 5 barrels/day be excluded from reporting under Subpart W or be
permitted to apply a default emission factor from the API Compendium in place of conducting
Response: EPA has included additional methodologies to determine operational parameters that
require sampling, namely low-pressure separator oil composition and Reid vapor pressure. For
further details, please refer to the response in EMAIL-0002-5 (comment also located in
The commenters interpretation of using engineering estimates to determine the operational
parameters for onshore production storage tanks is correct. This has been clarified in todays
final rule.
onshore production storage tanks. Please see the response to EPA-HQ-OAR-2009-0923-1061
10.
This section on onshore production and processing storage tanks requires use of the latest
software package for E&P Tanks. Section 98.7, however, references Version 2. It should allow
use of Version 2 or most recent version. In addition, the rule should allow use of commercially
available process simulators such as HYSIS or other similar process simulators. These are
industry accepted sophisticated process simulators used across the industry to design these units
and estimate emissions for permitting purposes. See TSD at 50. The appropriate parameters used
1434
to conduct process simulation may vary depending on site specific conditions and the list
included in this section should accommodate the variation.
Response: In todays final rule, EPA is allowing the use of alternate process simulators to
estimate emissions from onshore production storage tanks under certain conditions. For further
Commenter:

GPA recommends that HYSYS be identified as an acceptable method for calculating GHG
emissions from tanks, in addition to E&P Tanks. The use of HYSYS for tank calculations is
particularly important due to recognized difficulty in collecting acceptable separator liquid
samples when using E&P Tanks. In some case, the success rate for collecting acceptable
separator liquid samples can be as low as 10%. HYSYS provides the flexibility to use extend gas
analysis to model tank emission.
EPA has included additional methodologies to determine operational parameters that require
sampling, namely low-pressure separator oil composition and Reid vapor pressure. For further
details, please refer to the response in EMAIL-0002-5 (comment also located in rulemaking
memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923).
E&P Tank
We would like to see EPA include other approved flash simulation packages in their
methodologies for estimating carbon dioxide equivalent (CO2e) from this industry sector.
Some of these packages include, but are not limited to ProSim, HYSIM, HYSYS, ProMax,
KFlash, Vasquez-Beggs, Tanks 4.0. Utilization of one or multiple of these packages will
provide accurate flashing, standing, working and breathing (F/S/W/B) losses.
1435

Organization: Milan Steube
Commenter: Milan Steube
Subpart W as proposed requires the use of "E&P Tank Version 2.0 for Windows Software" to
determine field tank emissions for exploration and production operations. I'm not familiar with
this tool. Rather, EPA's TANKS 4.09d is commonly used for this purpose in the regulatory work
I do here in Southern California. Will I need to purchase and learn to use a new tool for the
purpose of reporting emissions under Subpart W?
Note to EPA: I am forwarding this question to you in case you want to add something regarding
why TANKS 4.09 cannot be used. I checked the TSD but it doesnt discuss TANKS. Thanks.
Response:
Once finalized, facilities subject to subpart W would be required to calculate and report GHG
emissions for all affected sources located at their facility using those calculation methods
specified in the finalized subpart W. No other calculation methods would be permitted.
Response: In regards to EPA TANKS model, please see response to EMAIL-0005-1 (comment
also located in rulemaking memo Early Comment Submissions in docket EPA-HQ-OAR
2009-0923) for clarifications on the use of this model for the purposes of Subpart W. However,
in todays final rule, EPA is allowing the use of alternate process simulators to estimate
emissions from onshore production storage tanks. For further clarification, please see the
For small combustion sources, the use of burner ratings and estimated operating hours rather
than direct fuel and/or operating hour monitoring to estimate emissions. For example:
* Especially during winter months, production fields employ numerous small heaters and other
combustion systems. Separators fire periodically to maintain temperatures necessary to separate
oil, water, and gas in the production fluid. The separator burner controllers are typically on/off
systems; thus when the burners fire they combust gas at the rated capacity. Operators understand
the annual duration of the separator burners operation and can estimate the number of hours of
combustion.
- For estimating flash gas emissions from production liquids storage tanks, process simulators
(such as HYSIS [superscript registered trademark symbol]), correlation equations (such as the
Vasquez-Beggs Equations), and other approaches have been accepted by state agencies for years
as accurate estimations of flash gas emissions and are appropriate for these estimates. Process
simulations can also provide estimates of CO2 dissolved in production liquids. For example it is
1436
not clear if E&P Tanks is appropriate for estimating emissions from produced water streams (as
required for produced water storage tanks by Section 98.233(j); if E&P Tanks is not appropriate,
then HYSIS [superscript registered trademark symbol] or an alternative process simulation
software appropriate for water streams would be required;
Response: In todays final rule, EPA does not require reporting of emissions from any external
combustion equipment with a rated heat input capacity equal to or lower than 5 mmbtu/hr from
reporting; however, an equipment count by type has to be reported. Please see Section II.F of the
preamble to todays final rule for more details. Also, EPA is allowing alternate methods to
determine the composition of field gas used in combustion equipment. Please see Section II.E in
the preamble to todays final rule for further details regarding changes to field gas composition
requirements.
clarification, please see the response to EPA-HQ-OAR-2009-0923-1061-10. Also, todays final
rule has been clarified to not include produced water storage tanks for onshore production.
Please see response to EPA-HQ-OAR-2009-0923-1305-48.
(Preamble p. 65) EPA seeks comment on whether there are additional or alternative software
packages to E&P Tank and GlyCalc that should be required to be used to calculate emissions.
The rule should allow reporters to use E&P Tanks and GlyCalc or any commercial process
simulator software to estimate emissions from tanks and glycol dehydration. The rule should also
not specify the version number of the software packages, as these programs are updated
periodically. In addition, in Section VII.I of this document, API outlines an alternative estimation
method for tanks than does not require modeling, but would provide accurate emissions
estimation while reducing burden.
EPA is allowing the use of alternate process simulators to estimate emissions from dehydrator
vents. For further details please see Section II.D of the preamble to todays final rule.

1437
Onshore Production and Processing Storage Tanks

Subpart W also requires oil and gas operators to monitor emissions from onshore production and
processing storage tanks using an engineering estimation based on the latest software package
for E&P Tank. 40 C.F.R. 98.233(j), 75 Fed. Reg. at 18640. IOGA-WV notes that the E&P
Tank program is designed for four specific tank systems: (1) low pressure gas separator, (2) high
pressure gas separator, (3) low pressure oil separator and (4) high pressure oil separator. If an
operator has an operating system that docs not fall within these four scenarios, however, E&P
Tank is not appropriate to use. IOGA-WV requests that the agency establish clear guidelines
outlining how operators should proceed in scenarios that fall outside of the four systems
identified above.
Response: EPA disagrees with the commenter that E&P Tank is designed for four specific tank
systems. E&P tank is designed to model two specific systems: tanks with separator, and stable
oil tanks (tanks without separator). The low-pressure gas, high-pressure oil, or low-pressure oil
are streams flowing into or out of the separator. In cases where an alternate operating system is
observed E&P Tank may have limited use. However, in todays final rule, EPA is not
constraining the reporters to the use of E&P Tank. For further information on alternate process
simulators to estimate emissions from onshore production and processing storage tanks; please
see the response to EPA-HQ-OAR-2009-0923-1061-10 for further details.
Subpart W in 40 CFR 98.233 refers to a software package for E & P Tank. Where can I find
more information on E & P Tank? Is this the EPA "tanks" program?
Response: The E & P Tank Version 2.0 software is available for purchase from IHS Standards
Store, Jane's Information Group, Inc., 110 North Royal Street, Suite 200, Alexandria, Virginia
22314 (http://www.ihs.com). The EPA Tanks program is a separate software that estimates
volatile organic compound (VOC) and hazardous air pollutant (HAP) emissions from fixed- and
floating-roof storage tanks. EPA Tanks is not applicable to todays final rule.

The proposed rule would require that emissions measurements be taken for all tanks, without
regard to size, throughput, and/or storage of volatile liquids, i.e_ actual potential to emit GHGs.
Scrubber dump valves on tanks can be operated either automatically or manually. When operated
manually, an operator is on-site and can determine if the valve has stuck and will immediately
1438
address the failure, so application of an emission factor to this scenario would greatly over
estimate the leak rate from this source. This requirement should be limited to only condensate
dump tanks with automatic dump valves, not all storage tanks.
Response: In todays final rule, EPA has included an equipment threshold for onshore
production storage tanks. For further details, please see the response to EPA-HQ-OAR-2009
0923-1061-10. Also, todays final rule has been clarified to not include produced water storage
tanks for onshore production. Please see response to EPA-HQ-OAR-2009-0923-1305-48.
EPA considered this comment and determined that reporters are required to report both
automatic and manual dump valves in todays final rule. EPA has provided procedures in todays
final rule to adjust for the time the dump valve has been open in the reporting period. Hence if
the dump valve is known to have not been open, then the calculation avoids any adjustment in
tank emissions.
98.233(j) Onshore Production and Processing Storage Tanks
a.) We interpret this paragraph to only apply to storage tanks receiving produced hydrocarbon
(crude oil or condensate) liquids. Our conclusion is based on:
The functionality of E&P Tank only applies to hydrocarbon liquids. The E&P Tank
software help page describes E&P TANK as a package the predicts hydrocarbon emissions
from oil production tanks. The description in the API Publication catalog describes it this
way: Publ 4697 Production Tank Emissions Model (E&P TANK, Version 2.0) E&P
TANK, developed in conjunction with the Gas Research Institute, is a personal computer
model designed to use site-specific information in a user-friendly format to predict
emissions from petroleum production storage tanks. The model calculates flashing losses
and simulates working and standing losses, using data provided by the user. Calculations
distinguish between HAPs and VOCs, showing detailed speciated emission rates from
methane to decanes.
In 98.233(j), the following parameters are used to characterize emissions:
(i) Separator oil composition.
(iv) Sales oil API gravity.
(v) Sales oil production rate.
(vi) Sales oil Reid vapor pressure
In 98.6, Sales oil means produced crude oil or condensate measured at the production
lease automatic custody transfer (LACT) meter or custody transfer meter tank gauge.
1439
EPA stated in the preamble in II. E. 2. (2) 4th paragraph on page 18621 that this proposed
rulemaking does not include emissions from tanks containing primarily water with the
exception of transmission station condensate tanks
We request that EPA clearly state that 98.233(j) Onshore Production and Processing
Storage Tanks does not apply to produced water tanks
b.) We request that EPA be clear in stating this category does not apply to flowback tanks used
for oil well testing. In our Alaska operations, open-topped tanks are used to flowback threephase flow (oil, water, gas) during oil well testing. E&P Tanks will not accurately estimate
such emissions. Rather, the GOR method proposed for the Well Testing and Venting
Category does a much better job.
c.) EPA should allow the operator to group tanks with similar operating conditions to establish
an emission factor (emissions/throughput) using E&P tanks for the group based on a
statistical sampling of tanks. The emission tank factor will be applied to the total number of
tanks to estimate the total annual emissions for all the tanks in the group.
Response: EPA agrees that the onshore production storage tanks does not include produced
water. With regards to clarifying the rule text under onshore production storage tanks, please
refer to EPA-HQ-OAR-2009-0923-1305-48.
Any emissions resulting from well testing have to be reported under the well testing venting and
flaring source type, which uses the GOR method.
In todays final rule EPA has provided an equipment threshold that considerably limits the
number of separators sending hydrocarbon liquids to tanks requiring detailed monitoring
methods (equipment below the threshold use a simplified emissions factor approach.) EPA
disagrees with commenter regarding allowing companies to develop emissions factors groups
tanks with similar operating conditions as EPA does not have sufficient data to provide standard
methodologies for statistical sampling of tank emissions nor did the commenter provide
suggested methodologies to estimate emissions factors. Hence todays final rule does not allow
for any sampling of tank emissions.
WCI Recommendations:
Onshore production and processing storage tanks (SECTION 98.233(j))
1440
USEPA requires the use of a modeling approach (E&P Tank) to calculate emissions from storage
tanks in the production and processing sectors. In cases where the model will not accurately
estimate emissions situations where additional emission sources are routed to the storage tank,
or vortexing is known to occur - USEPA requires reporters to apply a correction factor to
adjust emissions upward to account for these additional emissions. In the Technical Support
Document USEPA acknowledges several significant problems with this approach. The document
states that predicting and evaluating non-flashing effects on emissions (such as dump valves or
vortexing) has not yet been thoroughly studied or quantified. The methods above have significant
weaknesses as:
1. the sample data set (used to develop correction factors) is limited, and
2. only weak correlations were observed for the available data (r^2 = 0.0719 for oil and
0.0045 for condensate).
Thus, in its present form, this methodology would not produce cap-and-trade quality data.
Therefore, WCI recommends the use of a different methodology based on the measurement of
GOR for produced liquids at a location immediately prior to the storage tank. The inherent
assumption being that all gas contained in liquids entering a storage tank (at atmospheric
pressure) will be released to the atmosphere, either in the storage tank or as the liquid is
transported prior to refining. This method is presented below in the section titled WCI
Modifications.
Response: Because the Mandatory Reporting Rule has the primary intent to inform future
policy, EPA deems it inappropriate to fashion the MRR data reporting requirements with a
specific policy goal, such as cap and trade. In todays final rule, EPA has retained the factors
used to correct for additional emission sources routed to the storage tank; please see the response
to EPA-HQ-OAR-2009-0923-1305-48.
In todays final rule, EPA is allowing the sampling without modeling option. For further
clarification, please refer to the response to EPA-HQ-OAR-2009-0923-1305-48.
Onshore production and processing storage tanks
Section 98.233(j): The proposed rule requires modeling emissions from atmospheric pressure
storage tanks receiving produced liquids from onshore petroleum and natural gas production
facilities and onshore natural gas processing facilities. Since the preamble (page 18621)
explicitly states the proposed rulemaking does not include emissions from tanks containing
primarily water, IPAMS requests that EPA state in Section 98.233(j) that produced water tanks
are excluded. Accordingly, the activity data required for data reporting in Section
98.236(c)(10)(iv) should delete reference to water for liquids sent to atmospheric tanks.
1441
Response: EPA agrees that the onshore production storage tanks does not include produced
water. With regards to clarifying the rule text under onshore production storage tanks, please
refer to EPA-HQ-OAR-2009-0923-1305-48.
The EPA agrees with the commenters request to delete the references to water in Section
98.236(c)(10)(iv) in the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923
0002). In todays Final Rule, the text under the Section 98.236 for onshore production storage
tanks has been changed to only include sales oil.
AKA recommends that tank batteries ( 98.233(j)) with a throughput of less than 5 barrels/day of
hydrocarbon concentrate be excluded from reporting under Subpart W. If exclusion cannot be
provided, AKA recommends tank batteries below the 5 barrel/day threshold be allowed to use
the appropriate emission factor from the API Compendium of Greenhouse Gas Emission
Methodologies for the Oil and Gas Industry (API Compendium) in place of conducting
production storage tanks of 10 barrels per day. For further details, please see the response to
EPA-HQ-OAR-2009-0923-1061-10. However, the EPA is not using an emission factor from the
API Compendium for storage tanks below 5 barrels/day. Further details on the development of
the emission factor can be found in the rulemaking docket (EPA-HQ-OAR-2009-0923) under
Equipment Threshold for Tanks.

Another non-cost effective example of this is related to tanks, for which the same annual
calculation method is prescribed regardless of the tanks throughput, for example a tank that fills
at a rate of 1 barrel/day versus a tank that fills at a rate of 100 barrels/day. The prescribed method
requires analysis of the separator liquid and collection of numerous operating parameters. The
liquid analysis is expensive and often the sample is difficult or impossible to collect, even with
higher liquid production rates, and should not be required for tanks with minimal throughput.
Across the industry, there are thousands of tanks associated with very low liquid production rates
and, therefore, minimal GHG emissions.
production storage tanks of 10 barrels per day. For further details, please see the response to
EPA-HQ-OAR-2009-0923-1061-10.
1442
EPA has included additional methodologies to determine operational parameters that require
sampling, namely low-pressure separator oil composition and Reid vapor pressure. For further
details, please refer to the response in EMAIL-0002-5 (comment also located in rulemaking
Per the onshore production and processing storage tanks provision in 98.233(j), in order to run
E&P Tanks program, as mandated in this Subpart, to calculate estimated emissions venting from
tanks, a liquid sample must be taken off the separator to determine the oil composition and the
Reid vapor pressure, two of the minimum parameters required to characterize emissions. In the
Permian Basin, Pioneer has approximately 1,800 tank batteries consisting of at least 2
hydrocarbon tanks/battery. According to a reputable laboratory in Texas, an oil composition
analysis costs between $500 and $700, depending on the API gravity of the sample, and a Reid
vapor pressure analysis costs $125. In addition, the laboratory imposes a $180 charge for a
quality check analysis on each sample. For a laboratory to determine the oil composition and
Reid vapor pressure at all tank batteries in Permian, assuming there is only one separator/battery,
the laboratory analysis alone would cost, on the low end, approximately $1.4 million. This does
not include the time and labor to travel to each battery to collect the samples, or the cost to
purchase or rent the proper equipment to take samples. Also, if the battery has additional
separators, the costs would increase proportionately. Pioneer requests that one sample be taken in
each basin, or each "reporting unit" (per the reporting unit" recommendation in point 1).
Further, in order to obtain the CO2 and CH4 mole percent to determine the volumetric and mass
emissions, the calculation methodologies for a majority of the onshore petroleum and natural gas
production sources refer to 98.233 (u) and (v). Section (u)(2)(i) states "If you do not have a
continuous gas composition analyzer, then quarterly samples must be taken." Pioneer does not
have continuous gas composition analyzers at tank batteries; therefore, per this Subpart, quarterly
samples would be required, however it is very unclear as to where these samples must be taken.
Are quarterly samples to be taken somewhere in the basin or at each tank battery? Further, this
could potentially be problematic because sampling points are limited.
Pioneer requests clarification of EPA's justification for quarterly sampling. The composition of
the gas would not vary significantly from battery to battery so quarterly sampling would be
redundant and Pioneer contends that annual sampling would be adequate. In addition, Pioneer
requests that one annual sample be required per basin, or alternatively, in each reporting unit that
is determined based on similar battery characteristics or throughput (per "reporting unit"
recommendation in point 1).
Response: EPA has included additional methodologies to determine operational parameters that
require sampling, namely low-pressure separator oil composition and Reid vapor pressure. For
further details, please refer to the response in EMAIL-0002-5 (comment also located in
1443

Onshore production and processing storage tanks
A study prepared for the Texas Environmental Research Consortium measured emissions rates
from several oil and condensate tanks in Texas and developed average emission factors based on
direct measurement of vent gas flow rates 211 . The U.S. GHG Inventory mentions this study but
indicates that [b]ecause of the limited dataset and unexpected jumps in data points which can be
attributed to non-flashing emission affects, the United States decided that further investigation
would be necessary before updating the inventory emission factor 212 . The study found the
direct measurement approach to be the most accurate for estimating oil and condensate storage
tank emissions at wellhead and gathering sites; however, other, less accurate, approaches appear
to be much more commonly used.
EPA proposes the use of modeling (E&P Tank) to calculate emissions from storage tanks in the
proposed reporting rule. Yet the TSD acknowledges significant weaknesses with this approach.
213
Again, only a move towards more direct measurement (e.g., direct measurement of gas oil
ratios, vent gas, and flow rates) will reduce these uncertainties in the reported data.
Response: EPA agrees with the commenter that the most accurate method of quantifying
emissions is through direct measurement. However, EPA evaluated the cost burden of
conducting direct measurement at all onshore production storage tanks and found that the cost
increase is not justified by the increase in accuracy. As a result, EPA introduced several
methodologies, besides direct measurement, that balance accuracy and cost. Hence todays final
rule does not require direct measurement of emissions from tanks.
In todays final rule, EPA does not require quarterly sampling to determine the gas composition.
Please see Section II.F of the preamble to todays final rule for further details.
211
Hendler A., Nunn J., Lundeen J., McKaskle R., VOC Emissions from Oil and Condensate Storage Tanks Final
Report, April 2, 2009, available online at
http://files.harc.edu/Projects/AirQuality/Projects/H051C/H051CFinalReport.pdf.
212
Chapter 3, p. 3-47.
213
See TSD at 134.

1444
Section 98.233(j) Onshore production and processing storage tanks. The rule should allow the
use of calculation methods that are currently acceptable under state-level permitting and
reporting programs. Additional details on APIs suggested alternative methods for onshore
production and processing storage tanks are provided in Section VII.I of this document.
Response: In todays final rule, EPA has included some of the methodologies suggested by the
commenter. For further details, please see the response to EPA-HQ-OAR-2009-0923-1305-48

Paragraph (j) Onshore production and processing storage tanks outlines the software package for
calculating F/S/W/B losses. As stated in the definitions section of these comments, we request
the ability to utilize other software packages to calculate these emissions. This same comment is
applicable to paragraph (2)(i) and (4) of this section.
Paragraph (j)(3) which currently states Calculate emissions from liquids sent to atmospheric
storage tanks vented to flares as follows: should be modified to Calculate emissions from
liquids sent to atmospheric storage tanks where vapors are routed to flare(s) as follows:. In this
Subpart the definition of Vented emissions means intentional or designed releases of CH4 or
CO2 containing natural gas or hydrocarbon gas (not including stationary combustion flue gas),
including but not limited to process designed flow to the atmosphere through seals or vent pipes,
equipment blowdown for maintenance, or direct venting of gas used to power equipment (such
as pneumatic devices). which is not accurate in describing the flow of gas from atmospheric
storage tanks to a combustion device.
EPA agrees with the commenter on the applicability of the word vented for vapors routed to
flares. In todays final rule, vented has been replaced with sent.
Paragraph Section 98.233(j), 2, i - The phrase or to flare should be added to the end of the
sentence. Furthermore, a typographical error was noticed in the word, Beneficial.
1445
Response: EPA does not agree with the commenter. The downward correction for beneficial use
does not apply to flares. When tank vapors are sent to a flare they result in GHG emissions from
a flare and are not completely negated from the net balance of emissions. However, EPA has
determined that stating thermal control devices in that section is incorrect and has made edits to
todays final rule to reflect the correction.
Sections 98.232(c)(10) and Section 98.233(j): For each tank in hydrocarbon service, the rule
requires inventory of the tank physical and operational parameters, sampling (@ separator
pressure) and analysis of the inlet liquid stream, and modeling of emissions using E&P Tanks.
Since the sampling occurs at the separator, the cost comparison is based on the number of
separators, some 380,000 in natural gas and oil production operations from EPAs national
inventory. Assuming that the rule covers 81% of these, some 308,000 separators require
sampling under the rule. Assuming that the physical and operational inventory is embedded in
the well-site inventory and therefore zero incremental cost, that the compositional sampling and
analysis of the hydrocarbon stream from each separator upstream of a tank costs $475, and that
modeling each tank takes 0.5 hours at $100/hr, this yields a total cost of $192.3 MM.
Response: EPA disagrees with the commenters cost estimate. It was never EPAs intent to
require sampling of the low pressure separator oil. In todays final rule, EPA has allowed for the
use of default compositions available from the simulation software program or best available
data. See response to comment EMAIL-0002-5 (comment also located in rulemaking memo
Early Comment Submissions in docket EPA-HQ-OAR-2009-0923) for further details. Also,
EPA has provided equipment thresholds that significantly reduce burden by only requiring the
detailed monitoring methods for equipment above the threshold; equipment below the threshold
use emissions factors. Please see the response to EPA-HQ-OAR-2009-0923-1061-10 for further
details. These changes to todays final rule should significantly reduce burden to reporters.
Please see responses to EPA-HQ-OAR-2009-0923-1151-91 for further details on burden to
report from this source.
98.233 (j)(2)(i) The rule states that you can reduce the tank emissions calculated using EP Tank
if you have vapor recovery. Do you still need to run EP Tank even though you have vapor
recovery on the tank 100% of the time?
Response: In todays final rule, reporters must quantify emissions from tanks with vapor
recovery units regardless of whether the vapor recovery unit is operational throughout the
calendar year. EPA has determined that VRU systems are rarely operational throughout the
1446
calendar year and need to often be taken offline for maintenance. None the less, collecting data
on emission reductions will make it possible for EPA to understand how much of the total tank
emissions nationally are being captured to inform effective policy. Hence EPA has retained the
requirement on all emissions being estimated before making a reduction adjustment.
The Hy-Bon Engineering report should not be the basis for any EPA decision. At proposed
98.233(j)(4), relating to storage tank emissions calculations, a factor of 3.87 is prescribed for
sales oil < 45 API gravity. It appears that EPA chose this 3.87 factor based on a 2009 draft report
written for the Texas Commission on Environmental Quality (TCEQ) by a company called HyBon Engineering. 214 That draft report was thoroughly flawed and, as far as we know, was never
finalized and it was never used by TCEQ for any purpose. The many defects in the Hy-Bon
Engineering report were detailed in comments submitted to TCEQ by TPA in 2009, a copy of
which is attached to these comments. Accordingly, EPA should not base any decision in this
rulemaking on the contents of the Hy-Bon Engineering report.
Response: EPA does not agree that the TCEQ study cannot be used to derive the factors. EPA
studied TPAs critique of the TCEQ draft report by Hy-Bon Engineering, and while EPA has no
comment on the overall points that TPA is making about the qualifications or motives of Hy-Bon
Engineering, EPA did conclude that there is no basis for disputing the methodology or results of
tank direct emissions measurements. EPA also concluded that this is no basis for disputing the
flow rate of oil from the low pressure separator, the separator pressure during the measurement
period or the oil type. Therefore, while EPA considers it appropriate to question the cause of
excessive vapor emissions, EPA does not dispute the fact that in 50% of the tanks, the vapor
measured out the tank roof far exceeded the modeled flashing emissions. To EPAs knowledge,
neither TPA nor other peer reviewers of this report have offered any solid data or analysis
demonstrating how such a finding (much more vapor venting from a tank than can be accounted
for by an equilibrium flash calculation model) could be explained other than field gas from the
gas-liquid separator somehow bypassing the liquid level control valve. This TCEQ study finding
is corroborated by another independent study reported by HARC, which also found that many of
the tanks studied had far higher vapor emissions from the tanks than could be accounted for by
modeling the separator operating conditions and liquid properties for oil going into the tanks. For
further information, see the response to EPA-HQ-OAR-2009-0923-1305-48.
214
See EPA Technical Support Document at 51.

1447
This provision attempts to account for tank emissions associated with an equipment malfunction
event when a scrubber dump valve fails in the open position. There is an inherent weakness in
the calculation method for this section because it is based on empirical data with a very poor
correlation and very limited sample size. See TSD at 132. This section should be deleted, as
should any calculations for GHGs occurring during malfunctions.
Response: EPA does not agree that the TCEQ study cannot be used to derive the factors. For
further information, see the response to EPA-HQ-OAR-2009-0923-1305-48.

Commenter:
Comment on 98.233(j):
WBIH recommends the use of direct measurement on manifolded vent lines, existing and newly
constructed, in response to this measurement option, as an optional monitoring method for
emissions quantification for storage tanks at onshore petroleum and natural gas production
facilities.
Response: EPA has provided several monitoring method options in todays final rule. Direct
measurement by reporters was deemed onerous in terms of costs. Also, to keep the monitoring
methods consistent, EPA has not included the use of direct measurement for tank emissions.
Section 98.233(j) Onshore production and processing storage tanks
The rule indicates that this source applies to emissions from atmospheric pressure storage tanks
receiving produced liquids from onshore petroleum and natural gas production facilities
(including stationary liquid storage not owned of operated by the reported.)
The inclusion of stationary liquid storage tanks not owned or operated by the reporter is a
complex and unreasonable requirement as reporters would be required to gather technical
information, related to sales oil crude, from assets which they do not own or operate and this
information may be confidential. In many cases multiple owner/operators may feed one crude oil
storage tank; therefore, reporting agreements across multiple produces and field ownership will
be very complex and difficult to resolve. The boundaries of this reporting requirement are also
not clear, as BP feeds tanks not owned by BP and these tanks operate in series; therefore, BP is
unclear if they would report the first tank or all sequential tanks as well. BP recommends the
1448
reporter only be required to report emissions from oil storage tanks owned/operated by the
reporter.
Response: EPA disagrees that only emissions from onshore production storage tanks owned by
the reporter should be reported. Todays final rule requires gathering data from separators, which
are owned and operated by the reporter, to quantify emission from onshore production storage
tanks. However, reporters will have to collect information on vapor recovery systems from
owner and operators of onshore production storage tanks. For further details on legal authority
to require reporting from leased, rented, and contracted equipment, please see the response to
EPA-HQ-OAR-2009-0923-1031-21.
Section 98.233(j) Onshore production and processing storage tanks. The rule indicates that this
source applies to emissions from atmospheric pressure storage tanks receiving produced liquids
from onshore petroleum and natural gas production facilities (including stationary liquid storage
not owned of operated by the reported.) Application of this section is assumed to apply only
when the tank is owned or operated by the Sub-basin entity.
Response: EPA disagrees that the requirements to report emission from storage tanks not owned
or operated by the reporter should be removed. Please refer to the EPA-HQ-OAR-2009-0923
1305-23.
We also note that this provision refers to emissions from atmospheric pressure storage tanks.
Because storage tanks are either atmospheric or pressurized, but not both, the word pressure
should be deleted here and in 98.233(j)(1). Lastly, this provision would require emissions
reporting of tanks not owned or operated by the reporter. This requirement should be deleted
because it inappropriately creates obligations related to equipment over which owners and
operators have no legal rights or other responsibilities.
Response: EPA disagrees with the deletion of the word pressure from the phrase atmospheric
pressure storage tanks. EPA disagrees that the phrase cited by the commenter suggests that
atmospheric and pressurized tanks need to be reported. The phrase states that only storage tanks
at atmospheric pressure must report emission under Calculating GHG Emissions Onshore
Production Storage Tanks in todays final rule.
EPA disagrees that the requirements to report emission from storage tanks not owned or operated
by the reporter should be removed. Please refer to the EPA-HQ-OAR-2009-0923-1305-23.
1449
13.2.2.2
TRANSMISSION STORAGE TANKS

Tank Emission Requirements for Gas Transmission Compression Facilities Should Be Limited
to Condensate Tank Emissions and Clarifications are Required
98.233(k) identifies emission reporting requirements for storage tanks located at natural gas
transmission compression facilities. In the Proposed Rule preamble, EPA indicates that
atmospheric tank working and breathing emissions associated with the condensate are
insignificant. Rather, the concern is, scrubber dump valves malfunction or stick-open due to
debris in the condensate and can remain open resulting in natural gas bypass via the open dump
valve to and through the condensate tank. [75 FR 18620] EPA provides no indication that other
storage tank emissions for gas transmission are a significant emission source. However,
98.233(k) includes provisions for reporting from additional tanks.
INGAA recommends that 98.233(k) be limited to condensate tank emissions and deleting
98.233(k)(3), which references other tanks and calculation methods that apply for onshore
production and processing. EPA should rename the source as malfunctioning scrubber dump
valve venting from condensate tank to clarify the source of interest and avoid confusion with
emission estimates commonly applied for atmospheric tank emissions.
If tanks and emission sources other than condensate tank emissions from malfunctioning
scrubber dump valves are intended to be included, EPA should explain the significance of these
emissions and why they should be considered for natural gas transmission sources, per the
80/20 rule concerning insignificant emissions. In addition, calculation methodology should be
clarified with options other than the optical camera allowed for detecting emissions from a
malfunctioning scrubber dump valve.
A. Transmission Tank Emissions Reporting Should Be Limited to Condensate Tank Emissions
The Proposed Rule should be revised to appropriately refer to a malfunctioning scrubber inlet
dump valve as the intended GHG emission source. Although the measurement point is located at
condensate tank vents, defining transmission storage tanks as a source creates a broad
undefined source category that extends monitoring, measurement and reporting provisions
beyond the intended emission source.
The Proposed Rule preamble defines the five sources where EPA is requiring direct
measurement, including natural gas transmission storage tanks. EPA has characterized these five
sources as significant enough to warrant reporting, and the preamble states that no credible
engineering method or emission factor exists to accurately quantify emissions. EPA further
explains malfunctioning scrubber dump valves as the primary intended GHG emission source.
These valves are located upstream of the transmission condensate tank and are the focus of the
1450
required leak screening and measurement requirements. The intent is clearly reflected in several
quotes from 75 FR 18620 that are provided here:
EPA is proposing to require five sources in this supplemental proposal to directly measure
emissions: storage tanks (transmission) when scrubber dump valves are detected leaking
(emphasis added)
The preamble further discusses rationale surrounding the inclusion of transmission storage tanks
in direct emissions measurement. This discussion states that the actual transmission storage tank
condensate volume is typically low and emissions are considered insignificant, emphasizing the
intent to include the condensate tank only for the purpose of measuring emissions from
malfunctioning scrubber dump valves.
For example, storage tanks in the onshore natural gas transmission segment typically store the
condensate (water, light hydrocarbons, seal oil) from the scrubbing of pipeline quality gas. The
volume and composition of liquid is typically low and variable, respectively, in comparison to
the volumes and composition of hydrocarbon liquids stored in the upstream segments of the
industry. Hence the emissions from condensate itself in the transmission segment are considered
insignificant (emphasis added).
The preamble also acknowledges that the measurement requirement is directed at identifying and
measuring malfunctioning scrubber dump valves that have the potential to remain open, resulting
in unintended gas bypass through the condensate tank.
If the scrubber dump valve is stuck and leaking natural gas through the tank then the emissions
will be visibly significant and will not subside to inconspicuous volumes. If the scrubber dump
valve functions normally and shuts completely after the condensate has been dumped then the
storage tank, emissions should subside and taper off to insignificant quantities.
Since the preamble clearly identifies the intent to include identification and measurement of
malfunctioning scrubber dump valves, the Final Rule should limit transmission storage tanks to
condensate tank vents to capture the scrubber dump valve as the GHG emission source and only
refer to the associated tank vapor vent as the measurement point for determining excessive
leakage. INGAA is receptive to monitoring the condensate tank vapor vent to identify leaking
scrubber dump valves. To address this, INGAA recommends that 98.232(e)(3) be revised to
read as follows:
(3) Transmission condensate storage tanks venting due to scrubber dump valve malfunction.
Subsequent rule reference to this source should apply the same nomenclature. Without this
clarification, the undefined open-ended definition of transmission storage tanks could
significantly broaden this source category to insignificant tank sources that were not
characterized by EPA as emission points of concern and apparently not intended for reporting.
Per EPAs discussion in the preamble, the volume and composition of liquid are typically low
and variable from this segment and GHG emissions are expected to be insignificant.
Although INGAA strongly opposes inclusion of additional transmission compressor station
1451
storage tanks, if EPA decides to do so, the Final Rule should clearly identify and define the tanks
of interest. EPA should document why inclusion of additional tanks is warranted. In addition, the
Rule should allow flexibility in applying software tools and simulation programs to estimate
such emissions (e.g., E&P Tanks, HYSIS, etc.). INGAA notes that E&P Tanks is the prescribed
methodology for storage tanks at processing facilities.
B. Clarify Condensate Tank Vent Measurement and Emissions Calculation
A single annual snapshot optical screening and measurement (if continuous for more than 5
minutes) of condensate tank vent(s) does not consider the estimated time operating in a
malfunctioning mode. While 98.233 (k) does outline the volume and concentration
measurements, this section does not provide a calculation methodology for determining the
annual emissions from a malfunctioning scrubber dump valve routed to a transmission
condensate storage tank.
98.233(k)(2) uses the undefined term continuous to determine whether the transmission
condensate tank vapor vent should be measured with a meter. However, this terminology is not
clearly defined and the emissions calculation method is not provided, including the time basis for
estimating the duration of the malfunction event. These items should be clarified. For example,
98.233(k)(2) should indicate that continuous does not mean measurable over the 5-minute
monitoring time, but rather continuous and unabated over the 5-minute monitoring time.
Similarly, 98.233(k)(2) should provide a calculation method and time basis for the calculation.
INGAA recommends the following formula to calculate total annual emissions:
Ea = V x CH4 x T
Where:
Ea = Annual methane emissions in cubic feet at ambient conditions from tank vapor venting
resulting from a malfunctioning scrubber dump valves;
V = Measured tank vapor flow rate in cubic feet per hour;
CH4 = Mole or volume fraction of methane in tank vapor; and
T = Approximate duration of malfunctioning scrubber dump valves in hours during the year
(maintenance records or event logs can be used to bound the duration of the venting event).
C. Vent Emissions Detection Technology Should Not Be Limited to the Optical Gas Imaging
Instrument
Alternatives to the optical camera should be allowed for identifying a malfunctioning scrubber
dump valve and associated vent emissions.
EPA states in the preamble that The only potential option for measuring emissions from
scrubber dump valves is to monitor storage tank emissions with a gas imaging camera to
determine if the emissions become negligible when dump valves close (emphasis added). [75
FR 18620] Additional methods are available, as acknowledged by EPA in the Spring 2006 EPA
Natural Gas STAR Partner Update Technology Spotlight:
methane leaks through dump valves can be discovered in several ways: (emphasis added)
1452
Go to the slop tank and listen for excess gas flowing to or through the tank (the easiest method).
Look at the sight glass on the scrubber (if so equipped); if there is no liquid in the vessel, a leak
should be suspected.
Examine the throttling of gas through a leaking dump valve; a valve with frost indicates a leak.
Listen for a leak through a dump valve.
Use ultrasonic leak detection equipment with a touch probe to locate leaks through dump
valves.
Similar to comments in the fugitive emissions section above, INGAA requests that EPA broaden
the detection and measurement technology suite of tools to include all demonstrated technologies
and alternatives that are used in practice. This should include alternative tools for vent gas
measurement including calibrated bags and high-volume samplers (i.e., see Comment IV.C).
Limiting the technology choice to optical infrared cameras is unnecessarily restrictive and
precludes better and more efficient detection and measurement instruments.
Response: EPA agrees that it would be useful to reporters if EPA clarified that monitoring of
transmission storage tanks applies to condensate tanks only, and todays final rule has clarified
this.
EPA disagrees, however, that it should rename the source to malfunctioning scrubber dump
valve venting from condensate tank. EPA considered renaming the source per commenter
requests, but determined that it is consistent with other sources in the rule to list the point of the
emission, as opposed to the cause.
EPA disagrees that transmission storage tank venting emissions are insignificant and should not
be included in the rule. Please see response to comment EPA-HQ-OAR-2009-0923-1009-6.
Through EPAs Natural Gas STAR Program, EPA has learned that transmission storage tank
venting from malfunctioning scrubber dump valves can be a significant source at transmission
facilities. EPA used expert such that data will be gathered from this important source to inform
future policy.
EPA disagrees that E&P Tank would be appropriate to monitor transmission storage tank vented
emissions due to malfunctioning scrubber dump valves. E&P Tank relies of equations of state to
model the volume and composition of gas that flashes out of a liquid solution due to a change in
conditions, such as a pressure drop. It does not simulate emissions from a malfunctioning
scrubber dump valve, which is the focus of this source.
EPA agrees that the transmission storage tanks monitoring method should allow for reporters to
account for the fact that after repair the scrubber dump valve will stop leaking. It will improve
the accuracy of the emissions estimates and better inform future policy. Todays final rule
1453
allows reporters to calculate emissions from the beginning of the reporting period to the time the
leaking dump valve(s) is fixed.
EPA agrees that acoustic leak detection instruments utilizing the instrument manufacturers
quantification methods is an acceptable alternative to optical gas imaging cameras when
screening for leaking scrubber dump valves. EPA sought to offer an alternative to optical gas
imaging that has both an instrument manufacturer standard methods, and also will be available to
all reporters; and thus, an option for acoustic leak detection of the dump valve has been added to
the transmission storage tanks monitoring methods in todays final rule. However EPA disagrees
with the other suggested methods. Please see response to comment EPA-HQ-OAR-2009-0923
1039-15. The two alternative methods that rely on the human ear to listen for the sound of a leak
are unacceptable because this does not rely upon an instrument manufacturers method and does
not yield emissions quantification. In addition, visual observance of valves that are frosted does
not rely upon an instrument manufacturers method and does not yield emissions quantification.
While reporters may still choose to look through the sight glass prior to or in addition to
screening, not every scrubber will have a sight glass.
EPA disagrees with the comment to allow reporters to choose methods that are not specified in
subpart W. Please see Section II.L of the preamble to The Final Mandatory GHG Reporting
Rule (Final MRR), (40 CFR part 98). EPAs analysis of leak detection and measurement
methods is outlined in the Technical Support Document EPA-HQ-OAR-2009-0923-0027, and in
Section II.E of preamble to the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009
0923-0002).
EPA disagrees that calibrated bags or high volume samplers are appropriate alternatives to
meters for gas leaking through a dump valve into a transmission storage tank. EPAs experience
has demonstrated that the volumetric rate of emissions from this source may exceed the capacity
of a high volume sampler and calibrated bags.
Commenter: Kim Dang
Transmission Storage Tank Emissions. The proposed text of Subpart W does not clarify whether
the storage tanks to be monitored under the proposed 40 C.F.R. SECTION 98.233(k) are
condensate tanks or other types of tanks, such as diesel, amine, glycol, or produced water
tanks. 215 The preamble indicates that EPA intends for the methodology to apply to condensate
tanks; Kinder Morgan supports that interpretation, and requests that EPA clarify the text of the
rule accordingly.
More fundamentally, EPAs proposed methodology for quantifying emissions from transmission
storage tanks is problematic in concept and raises significant safety implications. The premise of
215
See Proposed Subpart W, 75 Fed. Reg. at 18,620.

1454
the methodology appears to be that scrubber dump valve malfunctions, which are very rare
events, can be reliably detected using optical imaging of the tank vent stack, and then quantified
using a meter. However, EPAs method fails to distinguish emissions that are attributable to a
valve malfunction from normal tank emissions caused by changes in temperature, pressure, and
liquid level. These working and breathing emissions, while minor, would result in observation of
continuous vapors using an optical gas imaging instrument. This, in turn, would require the
installation of a meter to measure extremely minor emissions and not the scrubber dump valve
malfunction indicated in the preamble. In addition, condensate storage tanks are frequently
located high off the ground and cannot be readily equipped with fall protection devices such as
railings or harnesses.
As an alternative to EPAs untested and potentially unsafe proposed methodology, Kinder
Morgan strongly recommends that EPA adopt the use of E&P Tanks software to estimate
transmission storage tank emissions. E&P Tanks is a widely used software tool within our
industry and is the method currently accepted by EPA for calculating emissions from condensate
storage tanks. The E&P Tanks program also can be expected to provide a more reliable emission
estimate than direct measurement, because it accounts for seasonal and daily changes in
temperature, pressure, and liquid levels that are not taken into account in the proposed Subpart
W methodology.
If EPA does not accept the use of E&P Tanks for transmission storage tanks, Kinder Morgan
believes tank vapors can be more accurately measured using calibrated bags or high volume
samplers than using meters. Accordingly, if EPA is going to require the direct measurement of
emissions at transmission storage tanks, Kinder Morgan requests that 98.233(k)(2)(i) be
modified to reference 98.234(c) and (d), in addition to 98.234(b).
In addition, Kinder Morgan believes that the proposed rule is unclear with regards to the
frequency of leak detection required for transmission storage tanks. Section 98.233(k)(1) refers
to SECTION 98.234(a)(1) for how the optical gas imagining is to be conducted. According to
SECTION 98.234(a): You must use the method described as follow to conduct annual leak
detection of fugitive emissions from all source types listed in SECTION 98.233(p)(3) (i) and
(q) Kinder Morgan requests that SECTION 98.233(k)(1) be added to the source types listed
in SECTION 98.234(a).
Response: EPA agrees that the rule should be clarified that monitoring of transmission storage
tanks applies to condensate tanks only, and has made this clarification in todays final rule.
EPA disagrees that it will be difficult to determine the difference between working and breathing
losses and a malfunctioning scrubber dump valve. EPA agrees that working and breathing losses
are indeed small and intermittent, and thus should easily be distinguished as viewed through an
optical gas imaging camera for five continuous minutes. A malfunctioning scrubber dump valve
that does not close properly will vent large volumes of gas that will continually blow out of the
tank, thus EPA has determined that the emissions targeted by this source are clear.
1455
EPA agrees that safety should be a primary concern of reporters when monitoring transmission
storage tanks; please see the response to EPA-HQ-OAR-2009-0923-1024-11. Of particular note,
EPA allows the installation of permanent flow meters and the use of acoustic instruments so
reporters can monitor these sources remotely and therefore never compromise safety.
Regarding the comments on E&P Tank, and calibrated bags and high volume sampler, please see
EPA agrees that the proposed rule is not clear that reporters must monitor transmission storage
tanks annually with an optical gas imaging instrument, and this has clarified this in todays final
rule.

Transmission storage tanks typically contain condensate liquids generated from the scrubbing of
pipeline quality natural gas (small amounts of natural gas liquids). In the Technical Support
Document, USEPA states the following the volume of condensate is typically low in
comparison to the volumes of hydrocarbon liquids stored in the upstream segments of the
industry. Hence the emissions from condensate itself in the transmission segment are
insignificant.
However, in an effort to quantify emissions which take place when a scrubber dump valve
becomes stuck and gas flows directly from the separator to the storage tank, USEPA requires that
reporters monitor storage tank vent emissions with a meter and measure storage tank gas
composition to quantify emissions. Reporter must use an optical gas imaging instrument to
monitor storage tank emissions for at least 5 minutes. If emissions are continuous (the
assumption being that the scrubber dump valve is stuck open), then they must use a meter to
measure emission flow rate from the tank. Reporters would use this flow rate and gas
composition to quantify emissions. The calculation thus assumes that emissions have been at this
level for the entire reporting period.
WCI has concluded that this methodology is subject to such large uncertainty that the resulting
data would not be cap-and-trade quality. As USEPA states, this emissions source is insignificant,
except in cases where a scrubber dump valve is stuck open.
This emission source might better be addressed by implementation of emission control
regulations at the state jurisdiction level. Jurisdictions might require that reporters monitor the
operational time of separator pumps to determine whether a separator/scrubber dump has
malfunctioned, resulting in elevated emissions. If the scrubber pump run times are excessive,
reporters would be then required to remedy the situation.
1456
Thus, WCI recommends that reporters not be required to estimate emissions from transmission
sector storage tanks.
Response: Because the Mandatory Reporting Rule has the primary intent of informing future
policy, EPA deems it inappropriate to fashion the MRR data reporting requirements with a
specific policy goal, such as cap and trade.
Todays final rule requires that transmission tanks found to be venting gas from malfunctioning
scrubber dump valves are measured with a flow meter and the emissions be extrapolated for the
period between the beginning of the reporting period and to the time if the valve was fixed. EPA
is satisfied that this data will be sufficiently accurate to inform future policy.
EPA disagrees that this source should not be included in the rule. Please see response to
Pre-amble-page 65, EPA seeks comment on whether there are additional or alternative software
packages to E&P Tank and GlyCalc that should be required to be used to calculate emissions.
CAPP supports flexibility in the use of process simulation software in the estimating of
emissions from storage tanks and glycol dehydrators. There is no practical reason why a reporter
should not be allowed to use a process simulation package such as Hysys to estimate emissions
should they choose to do so.
Response: With regard to storage tanks vented emissions, in todays final rule, EPA has
provided for the use of alternative software programs. Please see the response to EPA-HQ-OAR
2009-0923-1061-10.
With regard to dehydrator vent, in todays final rule, EPA has provided for the use of alternative
software programs. Please see the response to EPA-HQ-OAR-2009-0923-0053-4.
This section on transmission storage tanks is another attempt to account for tank emissions
associated with an equipment malfunction event when a scrubber dump valve fails in the open
position. See TSD at 51. This is not a normal mode of operation and, as such, should be fixed as
soon as possible when identified. Instead, the proposed rule would require an operator to allow
the malfunction to continue until a crew can be mobilized to the site to complete direct
1457
measurement of the emissions associated with the malfunction. This requirement is clearly not
appropriate and should be deleted.
Response: EPA disagrees with the comment on scrubber dump valves, please see response to
EPA-HQ-OAR-2009-0923-1061-10.
Section 98.233(k) Transmission storage tanks
Section 98.233(k)(1) refers to Section 98.234(a)(1) for how the optical gas imagining is to be
conducted. Section 98.234(a) says You must use the method described as follow to conduct
annual leak detection of fugitive emissions from all source types listed in Section 98.233(p)(3)
(i) and (q) Neither Section 98.233(k)(1) or Section 98.234(a) indicates the frequency for
performing the optical gas imaging for transmission storage tanks. BP assumes the optical gas
imaging is to be performed annually and not every time there is a release from the scrubber
dump valve. EPA should clarify the frequency in the rule.
Response: EPA agrees that clarification of the monitoring frequency was necessary in todays
final rule. Please see the response to comment EPA-HQ-OAR-2009-0923-1024-30.
13.2.3
DEHYDRATOR EMISSIONS ESTIMATES
Commenter:

The next example is related to gas dehydration units, which are required by the proposed Subpart
W to estimate annual emissions using the GLYCalc processes simulator. As with tanks, there are
thousands of very small dehydration units operating across the natural gas industry with minimal
GHG emissions. GPA recommends that dehydration units with a throughput of less than 10
million scf/day be excluded from reporting under Subpart W. If an exclusion cannot be provided,
GPA recommends dehydration units below the 10 million scf/day threshold be allowed to use the
appropriate emission factor from the API Compendium in place of conducting modeling using
Response: EPA agrees with the commenter on providing an equipment threshold, but EPA does
not agree on the 10 million scf/day limit nor does EPA agree on the proposal to exclude all
dehydrators below 10 million scf/day. . With regard to annual throughput threshold for
dehydration units, please see the response to EPA-HQ-OAR-2009-0923-1011-39.
1458

Glycol circulation pumps (e.g. a Kimray pump) may be natural gas driven. Thus there will be gas
emissions associated with the operation of these pumps which utilize natural gas. The spent
Kimray pump gas is usually dumped into the rich glycol stream, and thus will be flashed off
upstream of the reboiler if a flash tank is installed or in the absence of a flash tank, it will be
flashed off in the regenerator and vented through the still column to the atmosphere. If this is the
case, there is the potential for double-counting these pump emissions as part of the flash tank
or reboiler emissions and in the regulation section dealing with vented gas-powered pneumatic
devices and pumps. One way to address this issue would be to require reporting of Kimray pump
emissions using the pneumatic device methodology if the pump emissions are vented directly to
the atmosphere. If the Kimray pump gas is routed to the rich glycol stream, reporters would not
be required to meter Kimray pump gas consumption because these emissions would be captured
by measurements made at the flash tank or reboiler vent.
If all (or a portion of) dehydrator emissions are controlled, that is pump gas, stripping gas and
flash tank and still gas off-gas emissions are sent to a destruction device, they should not be
reported here to avoid double counting.
Measurement Options: WCI evaluated the following options:
[Table]
1459
Response: EPA agrees with the commenter that as per the April 2010 proposed rule, EPA-HQ
OAR-2009-0923 there will be double counting of emissions if dehydrators use natural gas driven
glycol circulation pumps. Hence, in todays final rule, EPA has provided a conditional reporting
of glycol pumps from the natural gas driven pneumatic pump venting emissions source. If the
reporter estimates emissions from glycol pumps under the dehydrator vent emissions source then
the reporter neednt report the natural gas driven pneumatic pump venting source as stand-alone;
if not, then the reporter has to estimate emissions from glycol pumps under the natural gas driven
pneumatic pump venting emissions source.
WCI Recommendations: WCI recommends the use of Option 2 as proposed in USEPAs
reproposed subpart W. The model does not significantly underestimate glycol dehydrator
emissions. It also avoids the problems associated with direct measurement and covers both
glycol and desiccant dehydrators. The model calculates emissions from Kimray natural gas
powered glycol circulation pumps. Thus Kimray pumps can be exempted from the metering
requirements in the pneumatics methodology.
1460
WCI Modifications: We recommend using the approach provided by USEPA without any
modification.
USEPA Subpart W Approach:
USEPAs re-proposed subpart W requires emissions be reported from the following:
1) dehydrator vent stacks without vapor recovery or thermal control devices,
2) dehydrator vent stack emissions sent to a flare or used as fuel in the regenerator firebox/ fire
tubes, and
3) vented emissions from desiccant dehydrators occurring during the refilling process
Quantification methods for these three emission sources are discussed below.
Uncontrolled dehydrator vent emissions
Reporters must calculate CH4 and CO2 emissions from uncontrolled dehydrator vent emissions
using the simulation software pack GRI-GLYCalc Version 4.0. The regulation requires that, at
minimum, the following inputs parameters must be supplied by the reporter and input to the
simulation software.
i. Feed natural gas flow rate

ii. Feed natural gas water content
iii. Outlet natural gas water content
iv. Absorbent circulation pump type
v. Absorbent circulation rate
vi. Absorbent type: (TEG), (DEG), (EG)
vii. Use of stripping natural gas
viii. Use of flash tank separator (and disposition of recovered gas)
ix. Hours operated
x. Wet natural gas temperature, pressure, and composition
Controlled dehydrator vent stack emissions
Reporters must use methods found in section (n) Flare Stacks to calculate CH4 and CO2
emissions resulting from destruction of dehydrator vent stack emissions.
Emissions resulting from the venting and filling of desiccant dehydrators.
Periodically, desiccant dehydrators are depressurized and vented when desiccant is replaced.
These venting emissions are to be calculated in the following manner:
Step 1 Calculate natural gas volume released at standard conditions
E [subscript S, n] = (H * D^2 * P * P [subscript 2] * %G * 365d/ yr) / (4 * P [subscript 1] * T *
1461
1,000cf /Mcf)
Where:
E [subscript S, n] = annual natural gas emissions at STP (Mcf)
H = height of dehydrator vessel (ft)
D = Inside diameter of the vessel (ft)
P [subscript 1] = Atmospheric pressure (psia)
P [subscript 2] = Pressure of the gas (psia)
P = pi (3.14)
%G = percent of packed vessel volume that is gas (void volume)
T = time between refilling (days)
Step 2 Calculate CH4 and CO2 volumetric emissions using calculations in paragraph (u) of this
section
Step 3 Calculate CH4 and CO2 mass emission using calculation methods in paragraph (v) of
this section
Response: In todays final rule, EPA has retained the methodologies for calculating GHG
emissions from dehydrator vent, but has also provided an equipment threshold for dehydration
units. For further details, please see the response to EPA-HQ-OAR-2009-0923-1011-39.

X. Proposed Rule Section 98.233(e) Dehydrator vent stacks.El Paso requests confirmation that
engineering estimates are acceptable as the operational parameters used to run the GlyCalc
model. In addition, El Paso requests that dehydrator units with a throughput of less than ten (10)
MMscf per day be exempted from the rule.
Response: EPA agrees with the commenter on using engineering estimation and providing an
equipment threshold, but EPA does not agree on the 10 million scf/day limit nor does EPA agree
on the proposal to exclude all dehydrators below 10 million scf/day. With regard to the use of
engineering estimates for calculations and annual throughput threshold for dehydration units,

X. Section 98.233(e) Dehydrator vent stacks
1462
Industry standard practices or current applicable regulations do not require continuous

monitoring of pressure, temperature, flow rates and circulation rates for sources such as
dehydrators and storage tanks. Typically, for a dehydrator, temperature and pressure at the
reboiler are the only parameters monitored. The proposed rule requires the following parameters
for characterizing emissions from dehydrators:
- Natural gas flow rate;
- Feed natural gas water content;
- Outlet natural gas water content;
- Absorbent circulation pump type(natural gas pneumatic/ air pneumatic/ electric);
- Absorbent circulation rate;
- Absorbent type;
- Use of stripping natural gas;
- Use of flash tank separator (and disposition of recovered gas);
- Hours operated; and
- Wet natural gas temperature, pressure, and composition.
It is El Pasos interpretation that engineering estimates can be used for the operational
parameters used to run the GlyCalc model since the rule does not specify metering and
monitoring for these parameters. El Paso requests confirmation of this interpretation. In addition,
El Paso requests that dehydration units with a throughput of less than 10 million scf/day be
excluded from reporting under Subpart W or be permitted to apply a default emission factor from
the API Compendium in place of conducting modeling using process simulations.
dehydrators below 10 million scf/day; EPA requires simpler methods for equipment below the
limit as opposed to exclusion. After considering several comments on burden to report, EPA
decided that an equipment threshold for dehydrators would reduce monitoring cost burden on the
industry. Hence, in todays final rule EPA requires that only dehydrators equal to or greater than
0.4 million standard cubic feet per day throughput monitor their emissions using a software
program; all dehydrators with throughputs below this value use an emissions factor. For further
details on equipment threshold decision, please see Section II F of the preamble to todays final
rule. For further details on equipment threshold determination analysis, please see rulemaking
docket EPA-HQ-OAR-2009-0923 under Equipment threshold for Dehydrators. In addition, it
was never EPAs intent to require metering of throughputs from equipment. Hence EPA clarifies
that engineering estimates of natural gas flow rate and hours of operation are sufficient for the
calculations. Todays final rule also allows the assumption of saturated gas if feed natural gas
water content is not available. Finally, todays final rule allows the use of default outlet natural
gas water content and default wet natural gas composition when outlet natural gas content and
wet natural gas composition are not available.

1463
2. Dehydrator Vent Stacks

Subpart W requires oil and gas operators to calculate emissions from dehydrator vent stacks
using an engineering estimation based on GlyCalc simulation software. 40 C.F.R. 98.233(e),
75 Fed. Reg. at 18638. IOGA-WV requests that the agency confirm that, as long as all of the
parameters specified in 40 C.F.R. 98.233(e)(l) are used, operators can utilize the full capability
of the GlyCalc software for purposes of the required emissions calculations (i.e., operators may
conduct the standard gas analysis rather than an extended analysis).
Response: EPA agrees with this comment. Hence, in todays final rule EPA allows the use of
any simulation software that conforms to the requirements in Section 98.233(e) of todays final
rule. For further details, please also see Section II.D of the preamble to todays final rule.
98.233 (c)(2):
CAPP would like clarity on the following statement in sub-section (2): Who decides what is a
similar pump model? Is this a role the EPA would take on and how would a facility go about
receiving a decision on a similar pump model?
To avoid double counting of dehydrator glycol pump emissions, CAPP recommends that an
exception to the rule for natural gas driven glycol circulation pumps (ie. Kimray) since these
emissions would already be captured under dehydrator vent gas emissions (98.233e).
Due to the difficulty and burden of maintaining an accurate log of the liquid pumped for each
pump on-site CAPP recommends as an alternative that facilities assume the pump is running at
the rated design flow for a specified amount of time.
Response: With regard to similar pump models, EPA requires the use of population emissions
factors for pneumatic devices and hence the issue is not relevant in todays final rule. Please see
Section II.F of the preamble to todays final rule for further details.
details.
98.233 (e)(3):
Based on feedback from industry members, CAPP does not believe the volume vented during
dehydrator (or any other) vessel opening and refilling is significant enough to track. This activity
1464
should only happen once per year and result in a small overall percentage of the basin level
emissions. CAPP recommends that an alternative approach be taken by the EPA is it wishes to
track these emissions.
Response: In todays final rule EPA has provided an equipment threshold on blowdown vent
stacks. Any physical volume between isolation valves that is less than 50 standard cubic feet
does not have to report in todays final rule; please see response to EMAIL-0002-8 (comment
2009-0923) for further details. Hence, any dehydrator associated volume that is less than the
equipment threshold does not have to report.
(98.233 (e)) The rule first exempts glycol units using thermal recovery from reporting dehydrator
vent stack emissions. But in section (2) it requires that these emissions be reported. What do we
do? Are the emissions exempt or should they be reported
Response: Please see the response to EPA-HQ-OAR-2009-0923-1305-34 for the treatment of
glycol units in 98.233 (e).

11. The rule should allow the use of new versions of software packages as they
come on the market. Proposed 98.233(e) prescribes the use of simulation software package
GRI-GLYCalc Version 4.0. It may be that, after the rule is finalized, a new version of that
package will be introduced onto the market. The new version should be available for use by
those who are covered by the rule. Accordingly, the reference to Version 4.0 should be amended
by adding "or later versions that become available". Such an amendment should be made at any
point in the rule where the use of specific products or software packages is prescribed.
Proposed 98.233(e) should also be amended to clarify how often an operator must
characterize emissions as prescribed in subparagraph (e)(1). In addition, 98.233(e) should
contain a threshold level below which an operator could use the API Compendium instead of
GLYCalc; TPA suggests that such a threshold be set at 10 million scf treated flow rate I day.
Finally, subparagraph (e)(3) should be deleted because the emissions covered by (e)(3) will
already be covered as a blowdown subject to the "blowdown vent stacks" provisions set forth at
75 Fed. Reg. 18640 (those provisions are currently mis-numbered as "iii" in the middle column
of page 18640, as noted below).
1465
Response: EPA agrees with the comment about software programs. With regard to flexibility of
using software programs, please see the response to EPA-HQ-OAR-2009-0923-1014-12.
EPA agrees with the commenter on providing an equipment threshold, but does not agree on the
suggested limit. For further details, please see the response to EPA-HQ-OAR-2009-0923-1011
39.
EPA clarifies that the reporter must characterize dehydrator emissions only once in a reporting
period. Finally, EPA has clarified in todays final rule that desiccant dehydrators must report
under the dehydrator vent source type. Please see Section 98.233(e) of todays final rule for
further details.
Commenter:
Comment: WBIH supports the use of GlyCalc for dehydrator vent stacks emissions calculations
Response: In todays final rule, EPA has widened the calculation options by allowing the use of
other simulation software. For further details, please see Section II.D of the preamble to todays
final rule.

AKA recommends that dehydration units with a throughput of less than 3 million scf/day be
excluded from reporting under Subpart W. If an exclusion cannot be provided, AKA
recommends dehydration units below the 3 million scf/day threshold be allowed to use the
dehydrators below 3 million scf/day. With regard to annual throughput threshold for dehydration
units, please see the response to EPA-HQ-OAR-2009-0923-1011-39.

Another example is related to gas dehydration units, which are required by the proposed Subpart
1466
W to estimate annual emissions using the GRI GLYCalc processes simulator. As with tanks,
there are thousands of very small dehydration units operating across the natural gas industry with
minimal GHG emissions.
Response: EPA agrees with the commenter. With regard to GHG emissions from small
dehydration units, please see the response to EPA-HQ-OAR-2009-0923-1011-39.

E. Section 98.233(e) Dehydrator vent stacks
For well-site dehydrators, Subpart Ws requirement for engineering estimation is in direct
contradiction with the Technical Support Document (TSD). By following Figure 1: Decision
Process for Emissions Source Selection (TSD, p.22), the decision to use a population emissions
factor and source count to estimate emissions is determined through the decision tree as follows:
1. Is source contributing to top 80% of emissions from each segment? YES
2. Is emissions source a fugitive? NO
3. Is vented emissions source geographically dispersed? YES
4. Does credible emission factor exist? YES
5. Use population emissions factor and source count to estimate emissions.
API recognizes that some field dehydrators are already subject to GLYCalc runs under current
NESHAP Subpart HH. For these dehydrators, the incremental effort to calculate greenhouse gas
emissions is manageable. For the other geographically dispersed field dehydrators which have
less than 3 mmscfd actual annual average throughput and are not already required to perform a
GLYCalc analysis, we propose the option of using the API Compendium emission factors (in
Tables 5-1 and 5-2 for glycol dehydrator emissions) in lieu of direct measurement of parameters
for the Engineering Estimation method. This option is consistent with the TSD decision tree
outcome above.
As an alternative estimation option, API supports EPAs selection of modeling emissions from
dehydration units and supports the use of GRI GLYCalc or other commercial software process
simulators. API has several comments to clarify this category in the rule along with several
proposals to broaden the methodologies available and to streamline the determination of GHG
emissions from dehydration units thus significantly reducing the burden imposed by the
proposed rule. In addition, EPA should recognize that CO2 emissions are presented in the
regenerator overhead stream results from GLYCalc, not the emission stream.
As proposed, the rule restricts the selection of methods to modeling using GRI GLYCalc Version
4 and calculating controlled emissions separately. API believes the allowable methodologies
should be broadened as follows:
1467
a. Allow the use of GRI GLYCalc, Version 4 or higher.

b. Allow for calculating the flare or regenerator controlled emissions in GLYCalc.
c. Allow for the use of other commercial process simulators.
To address the dilemma of having to sample and model each individual dehydration unit and the
resulting burden and cost, API suggests an alternative approach which will significantly reduce
this burden while still yielding high quality data.
Dehydration Unit Calculation for a Sub-basin Entity
In a particular Sub-basin entity within a natural gas basin, the natural gas production from the
same or similar producing horizons/formations is very similar in composition, properties, and
behavior and does not vary widely from one well to another. In addition, for a given producing
well/site and natural gas composition, the lean glycol flow rate is the largest determinant of
emissions from the dehydration unit, including CO2 and Methane. These characteristics allow
for simplifying the emission estimation from glycol dehydration without sampling and modeling
each dehydration unit. The only information that will be required for each individual dehydration
unit is the lean glycol flow rate determined by pump strokes per minute, if a gas assist pump or
from pump setting information if using an electric pump. API has developed an example of how
this could be structured and could discuss this approach further with EPA.
Response: Upon further analysis and review, EPA has determined that a population emission
factor will provide acceptable emissions and activity data for smaller glycol dehydrators.
Therefore, todays final rule provides a threshold of 0.4 mmscfd for application of a population
emission factor derived using the 2009 API Compendium production factor. For further
information on this threshold, see the response to comment EPA-HQ-OAR-2009-0923-1011-39.
For glycol dehydrators above this threshold, EPA disagrees with the commenters application of
the decision tree approach. The decision step 4 Does credible emission factor exist? was
answered No. The single emission factor presented in the API Compendium correlates
emissions from all dehydrators simply with throughput. While this factor was derived using
GlyCalc, EPA has determined that emissions correlate better with glycol circulation rate,
contactor pressure and other attributes of the facility such as type of circulation pump, flash tank
separator and stripping gas injection. Each larger unit will be different, so using GlyCalc will
account for the lower emissions of operators who have installed emission reduction technologies,
whereas the API Compendium factor will treat all larger glycol dehydrators the same, except
throughput. Hence, EPA has retained the requirement for simulation of dehydrator emissions in
todays final rule.
EPA agrees with the comment about software programs. With regard to flexibility of using
software programs, please see the response to EPA-HQ-OAR-2009-0923-1014-12.
1468
EPA disagrees with the use of GLYCalc calculations for flare or regenerator controlled
emissions. For further details, please see the response to EPA-HQ-OAR-2009-0923-1305-52.
EPA disagrees with the alternative approach for a dehydration unit calculation for a sub-basin
entity suggested in this comment. For further details, please see the response to EPA-HQ-OAR
2009-0923-1305-52.
22. Dehydrator Vents
We support EPAs proposed emission estimating method for dehydrator vents that relies on the
GLYCalc model and engineering calculations for desiccant dehydrators.
Response: In todays final rule, EPA has retained the methodologies for calculating GHG
emissions from dehydrator vent and has also provided an equipment threshold for dehydration
units. EPA has also widened the calculation options by allowing the use of other simulation
software. For further details, please see the responses to EPA-HQ-OAR-2009-0923-1011-39 and
EPA-HQ-OAR-2009-0923-1014-12.
This section for dehydrator vent stacks should allow the use of GRI-GLYCalc Version 4.0 or
later. Also, it should establish a de minimus threshold below which the unit is excluded from
reporting or is allowed to use API Compendium to estimate emissions. There are thousands of
very small dehydration units operating across the natural gas industry with minimal GHG
emissions. GPA recommends that dehydration units with a throughput of less than 3 million
scf/day be excluded from reporting under Subpart W. If an exclusion cannot be provided, GPA
recommends dehydration units below the 3 million scf/day threshold be allowed to use the
In addition, for units over 3 mmscf/d should allow use of commercially available process
simulators such as HYSIS or other similar process simulators. These are industry accepted
sophisticated process simulators used across the industry to design these units and estimate
emissions for permitting purposes. See TSD at 43. The appropriate parameters used to conduct
process simulation may vary depending on site specific conditions and the list included in this
section should accommodate the variation.
Response: EPA agrees with the comment about software programs. With regard to flexibility of
using software programs, please see the response to EPA-HQ-OAR-2009-0923-1014-12.
1469
EPA agrees with the commenter on providing an equipment threshold, but does not agree on the
suggested limit as well as exclusion. With regard to annual throughput threshold for dehydration
units, please see the response to EPA-HQ-OAR-2009-0923-1011-39.

K. Section 98.233(e) Dehydrator vent stacks
BP supports EPAs selection of modeling as the methodology for determining emissions from
dehydration units and supports the use of GRI GLYCalc as an applicable model. However, BP
has several comments to clarify this category in the rule along with several proposals to broaden
the methodologies available and to streamline the determination of GHG emissions from
dehydration units thus significantly reducing the burden imposed by the proposed rule.
1. As proposed, the rule restricts the selection of methods to modeling using GRI GLYCalc
Version 4 and calculating controlled emissions separately BP believes the allowable
methodologies should be broadened as follows:
a. Allow the use of GRI GLYCalc, Version 4 or higher.
b. Allow for the calculating the flare or regenerator controlled emissions in GLYCalc.
2. To address the dilemma of having to sample and model each individual dehydration unit and
the resulting burden and cost, BP suggests an alternative approach which will significantly
reduce this burden while still yielding high quality data.
Dehydration Unit Calculation for a Reporting Unit/Area
In a particular producing area (Sub-basin entity) within a natural gas basin, the natural gas
production from the same or very similar producing horizons/formations is very similar in
composition, properties, and behavior and does not vary widely from one well to another. As
EPA is aware, for a given producing well/site and natural gas compositions, the lean glycol flow
rate is the largest determinant of emissions from the dehydration unit, including CO2 and
Methane. This characteristic allows for only acquiring the average gas composition, pressure,
and temperature within a Reporting Unit/Area and individual dehydration unit lean glycol
circulation rate to be used to develop a robust methodology of quantifying emissions from
individual dehydration units without sampling and modeling each dehydration unit. The only
information that will be required for each individual dehydration unit is the lean glycol flow rate
determined by pump strokes per minute if a gas assist pump or from pump setting information if
using an electric pump. This would be structured as follows:
1470
1. Split out dehydration units with flash tanks from those without flash tanks.
2. Based on at least 5 gas analyses, determine a weighted average composition (concentration on
a volume%, dry basis) based on the production rate for the reporting unit/area.
3. Determine the average pressure (psig) and temperature (degree F) for the reporting unit/area.
(For example, 365 psig and 75 degrees F based on field production data)
4. Determine the average water content of the dry gas (For example, 5 lbs H2O/MMscf for sales
gas requirements)
5. Determine the average water content of the lean glycol. (For example, 1.5 wt% H2O from
manufacture data)
6. Determine the maximum and minimum lean glycol flow rate (circulation rate) of each type of
glycol pump used.
7. Determine if a gas injection or electric/pneumatic pump is used.
8. Determine if flash tanks are used and if they are the average temperature and pressure. If there
are no flash tanks or flash tank vapors are vented to the atmosphere or used as stripping gas, use
the no flash tank option. If there are some flash tanks that are controlled or recovered, do a
separate set of runs at each of the lean glycol flow rate. Be sure to add the emissions from the
flash tank control device to the regenerator vent emissions.
9. Determine if stripping gas is used and if they are the type of stripping gas and the flow rate of
stripping gas (scfm) for each individual dehydration unit.
10. Determine if control devices are used. If a condenser is used determine the average
temperature and pressure. If a combustion device is used determine the average ambient air
temperature, excess oxygen, and destruction efficiency.
11. Using GRI-GLYCalc calculate the emissions from the dehydration unit at the maximum,
minimum, and one or two levels in between of the lean glycol flow rate per pump type (based on
manufacturer information) holding the following information constant:
- The average pressure and temperature for the reporting unit/area.
- The average composition (concentration by volume %, dry basis for the ) for the reporting
unit/area
- The average water content of the dry gas for the reporting unit/area
- The average water content of the lean glycol for the reporting unit/area.
12. Vary the lean glycol flow rate which with each run (0.33, 0.16, and 0.1 gpm)
1471
- Use the electric/pneumatic pump option even if there is a gas injection pump if there is no flash
tank or the flash tank emissions are uncontrolled. (Gas injection pump emissions will be added
separately by dehydration unit). If there is a flash tank that tank that is controlled and a gas
injection pump, use the gas injection pump option and enter in the calculated gas injection pump
volume ratio for the pump using the following formula from the GRI GLYCalc Glycol Pump
dialog box:
PVRa = PVRs * Ps * Ta / (Pa * Ts)
where:
PVRa, Actual Pump Volume Ratio (acfm/gpm)
PVRs, Standard Pump Volume Ratio (scfm/gpm) =
Pa, Operating Pressure (psig) =
Ta, Operating Temperature (oR) =
Ps, Standard Pressure (psia) =
Ts, Standard Temperature (oR)
- If there are no flash tanks or flash tank vapors are vented to the atmosphere or used as stripping
gas, use the no flash tank option. If there are some flash tanks that are controlled or recovered, do
a separate set of runs at each of the lean glycol flow rate. Be sure to add the emissions from the
flash tank control device to the regenerator vent emissions.
- Run with no stripping gas even if stripping gas is used by the various dehydration units.
(Stripping gas emissions will be added separately by dehydration unit.)
13. Save the file and run the calculations.
14. Print and save the Aggregate report. Note that the CO2 emissions are presented in the
regenerator overhead stream results, not the emissions stream.
15. Rerun the calculations at each of the different lean glycol pump rates.
16. Plot the emissions versus the flow rate for methane and carbon dioxide in Excel using the
Scatter Plot option. (Note the carbon dioxide emissions are not presented in the emissions
summary but must be pulled from the regenerate vent results.)
18. On the Type tab, select Linear.
19. On the Options tab, select Display equation on chart and Display R-squared value on
chart, then select OK,
20. Use the produced equation to calculate the emissions for each dehydration unit based on the
lean glycol flow rate excluding emissions from gas injection pumps if no flash tank or
1472
uncontrolled flash tank and stripping gas.

21. If a gas injection pump is used:
- Obtain from the manufacturer information the natural gas usage (scf/min) per glycol pump
(gpm) based on the operating pressure (linear function).
- Multiply the natural gas usage (scf/min) at the average reporting unit/area pressure by the
pumps lean glycol flow rate (gpm) to determine the natural gas emitted by the gas injection
pump.
- Using the average natural gas composition for the reporting unit/area, determine the emissions
of methane and carbon dioxide from the gas injection pump.
- Add the gas injection pump emissions to the rest of the dehydration unit emissions for both
methane and carbon dioxide.
22. If dry gas or flash gas is used as a stripping gas:
- Obtain the gas flow rate (scfm) for each dehydration unit.
- Using the average dry gas or flash gas composition, determine the methane and carbon dioxide
added by the stripping gas.
- Add the stripping gas emissions to the rest of the dehydration unit emissions for both methane
and carbon dioxide.
Response: In response to the first section of this comment, with regard to flexibility of using
software programs, please see the response to EPA-HQ-OAR-2009-0923-1014-12.
With regard to flare or regenerator controlled emissions, EPA has provided a method to calculate
these emissions from dehydrators in todays final rule; please see Section 98.233(e) of todays
final rule for further details. To keep the method consistent across the rule, EPA does not allow
the use of GLYCalc calculations for flare or regenerator controlled emissions.
EPA reviewed the commenters input and proposed alternative to the monitoring method
provided by EPA in the proposed rule. However, EPA has determined that providing an
equipment threshold, will significantly reduce the count of dehydrators that will need monitoring
using a software program and achieve the same result, i.e. burden reduction, as using an
alternative monitoring method suggested by the commenter . In addition, EPA has determined
that if steps 16-19 in the commenters proposed method are used to fit an equation to the
sampled data, it is possible to generate fits that are not reasonable, i.e. with adjusted R-square
values that are very low. In such cases the equation will not correctly represent the sampled
values and cannot be used to characterize the entire dehydrator population. As such, EPA has not
accepted the proposed alternative and instead provided relief in terms of an equipment threshold.
For further information, please see the response to EPA-HQ-OAR-2009-0923-1011-39.
1473
13.2.4
ACID GAS REMOVAL VENTS
Commenter:

GPA further recommends that ProMax be identified as an acceptable method for calculating
GHG emissions AGR units. In addition, GPA recommends that other mass balance calculations
be allowed for AGR units. For example, an operator typically has metered volumes and
composition analysis of gas entering an AGR unit and can conservatively assume 100% of the
CO2 is removed (in some cases a few percent of the CO2 may not be removed). A simple
calculation will provide accurate CO2 emissions from the AGR unit without the need for
downstream metering or analysis. In other cases, operators may have meters and analysis on the
AGR vent line and can provide a direct calculation of the CO2 emissions, again without the need
for downstream metering or analysis. Subpart W should provide flexibility for any of these
methods to estimate the GHG emissions from AGR vents.
Response: EPA agrees with the comment and has included additional options to estimate
emissions from AGR units. With regard to acceptable methods for calculating GHG emissions
from AGR units, please see the response to EPA-HQ-OAR-2009-0923-1024-26.

Measurement Options: Four options were evaluated for the calculation of ARU vented
emissions. These options are shown below:
1474
WCI Recommendation: WCI recommends use of the mass balance approach of USEPA (Option
4) with minor modifications. This method will provide cap-and-trade quality data, it is applicable
to multiple sour gas treatment technologies, and does not require additional sampling as the
required data is currently generated for process control purposes.
WCI Modifications: Language should be added to exempt reporting for this emission source in
cases where acid gases are re-injected into the oil/gas field.
USEPAs methodology calculates annual CO2 emissions based on either continuous gas
analyzer results or quarterly gas samples. Reporters must calculate and input the volume
weighted annual average CO2 content for the gas entering and exiting the AGR unit. WCI
recommends increasing the gas sampling frequency (determination of %Vol1 and %Vol2) to
monthly where a continuous gas analyzer is not installed. The increased sampling frequency
should not result in significant additional burden and should provide more accurate annual
emission data.
Response: EPA disagrees with the modifications suggested by the commenter. In the final rule
establishing the GHG Reporting Program (74 FR 56260, October 30, 2009), EPA was clear that
subpart methods and calculation procedures must be followed whether or not there is subsequent
injection underground or geologic sequestration. The GHG Reporting Program is not an
emissions inventory; rather it is a reporting program that collects data to inform future climate
change policies. The same rationale applies to subpart W in this final action. Data on CO2 from
1475
an acid gas recovery unit is needed by EPA to inform future climate change policies, even if the
CO2 stream is subsequently injected underground. Therefore, such CO2 streams must report for
the AGR unit emission source.
In cases where CO2 streams are injected underground for geologic sequestration, EPA provides
methods for reporting that information in the proposed subpart RR, Geologic Sequestration.
EOR operations may choose to opt-in to the proposed subpart RR.
This final rule will compile information on all emissions from sources in Subpart W, including
capture or reuse of GHGs. In the context of this comment, not all of the CO2 or acid gases
separated in the AGR unit may be captured for sequestration or EOR operations, i.e. the
efficiency of capture is not 100 percent. The difference between acid gases separated at the AGR
and sequestration reported under Subparts RR and W will allow EPA to determine true
emissions; an altogether exemption of AGR units that do a portion of the separation and capture
will not account for these emissions. Hence EPA has retained the requirement for all acid gas
recovery units to report.
In regards to the sampling, EPA determined that the incremental increase in accuracy from
increased sampling does not justify the incremental burden from increased sampling. Since the
intent of this final rule is to inform a range of policies and not provide data for compliance with a
specific program such as cap-and-trade, EPA has retained the requirement for quarterly
sampling.
Section 98.233(d) Acid gas removal (AGR) vent stacks
The proposed rule requires the following parameters for characterizing emissions from AGR:
- Metered natural gas inlet volume;
- Metered natural gas outlet volume;
- Volume weighted CO2 content of inlet gas;
- Volume weighted CO2 content of outlet gas.
Industry standard practices or current applicable regulations do not require continuous
monitoring of natural gas flow and volume weighted CO2 content of the gas in and out of the
AGR. Typically, for an AGR, natural gas inlet volume is the only parameter monitored. El
Pasos recent cost estimate indicates it would cost more than $115K to install a metering facility
to measure the inlet and outlet volumes for one amine unit. Given this high cost, an option to use
estimated natural gas inlet and outlet volumes should be allowed.
1476

Proposed Rule Section 98.233(d) Acid gas removal (AGR) vent stacks.El Paso encourages the
EPA to allow reporters to estimate the inlet and outlet volumes if not currently metered.
Response: EPA agrees with the comment and has allowed engineering estimation, under certain
conditions, to calculate emissions from AGR units. With regard to acceptable methods for
calculating GHG emissions from AGR units, please see the response to EPA-HQ-OAR-2009
0923-1024-26.
Acid gas emissions are calculated, not directly measured, using a mass-balance approach: Inlet
gas is measured as is outlet gas the difference is considered to be acid gas. This methodology
has been an accepted method with regulatory agencies. Gas that is high in H2S, is highly
corrosive: flow meters for acid gas have a high rate of failure and are highly unreliable due to the
highly corrosive environment. PAW proposes an additional monitoring methodology using the
mass-balance approach to calculating acid gas.
Commenter: Kim Dang
Acid Gas Removal Units. Individualized metering of the volume of gas flowing both into and out
of every acid gas removal (AGR) unit is not necessary to determine emissions and should not be
required. 216 Most AGR units at our facilities do not currently have meters directly upstream and
downstream of the units. To install meters at every unit would entail substantial cost and cause
considerable disruptions at our operations, since units would have to be shut down during meter
installation. In lieu of direct metering, or as an additional quantification option, Kinder Morgan
recommends that EPA allow appropriate computer modeling of AGR units, as the agency
216
Proposed 40 C.F.R. SECTION 98.233(d).

1477
proposes for emissions from onshore production and processing storage tanks 217 and dehydrator
vent stacks. 218 There are several commercially available and widely used software packages,
such as ProMax, that could be used to estimate emissions from AGR units with acceptable
accuracy. Given the expense of direct metering and the logistical and safety issues it would
cause, there appears to be no compelling reason why EPA should not permit the use of such
emissions estimation software.
Alternatively, Kinder Morgan suggests that EPA clarify the proposed methodology to allow
appropriate engineering estimation of volumes flowing through AGR units. In many cases, the
volume of gas entering and exiting an AGR unit can be derived from data recorded at existing
meters at other points in the facility. EPA should permit reporting entities to use such methods
where appropriate, rather than require direct metering at every AGR unit. 219
In addition, Kinder Morgan believes that Equation W-4 is difficult to apply to AGR units in EOR
operations that capture flashed CO2, compress the CO2, and inject it into the ground. As written,
Equation W-4 would classify this recycled CO2 gas as part of the units overall emissions even
though the CO2 is never released to the atmosphere. Appendix A shows how Equation W-4 can
be modified to reflect this use of CO2 at AGR units in service at EOR operations.
Response: EPA agrees with the commenter on the use of engineering estimation methods to
determine flow rate when no meters (either flow rate or vent stack) or CEMS are already in
placed. Typically, all of the gas processed in a facility is sent through an AGR unit to make the
gas pipeline quality by removing acid gases. Since the gas into the facility is metered at various
points, an engineering estimate can provide reasonable information on the gas throughput of the
AGR unit. EPA also agrees with the commenter that either the inlet or outlet feed gas volume
should be sufficient to reasonably estimate emissions. In todays final rule, EPA has provided
flexibility in determining the flow rate to the AGR units, since the inlet and outlet flow rates
differ by the amount of acid gas separated, and can be estimated using a mass balance approach.
This method of estimating emissions will provide reasonable accuracy required for Subpart W.
EPA also agrees with the flexibility to use simulation software where applicable. EPA allows for
the use of any simulation software that meets the conditions specified in 98.233 (d) (4).
Todays final rule requires the use of CEMS data on emissions where CEMS are already in
place. If CEMS are not already in place and not installed for purposes of this rule, data from
meters on vent stacks can be used. If meters on vent stacks are also not available or not installed
for the purposes of this rule then the reporter is required to use either any upstream or
downstream meter already in place or simulation software. If none of the previous options are
applicable, the reporter has the choice of adopting one of the previous options or using an
engineering estimation of the flow rate and calculation method to estimate emissions. Hence,
217
Proposed 40 C.F.R. SECTION 98.233(j).
218
Proposed 40 C.F.R. SECTION 98.233(e).
219
If necessary, EPA can require that reporting entities keep documentation explaining and justifying the
engineering calculations used to estimate volumes entering and exiting an AGR unit.
1478
EPA has provide sufficient flexibility to use multiple options from the most accurate to the most
reasonable of monitoring methods as applicable to different facilities.
EPA understands that AGR units in EOR operations may recycle all or a portion of their
captured CO2. If a an EOR operation chooses to opt-in to the proposed subpart RR and report
geologic sequestration, EPA will account for the recycled CO2 as part of the proposed subpart
RRs methodology for quantifying geologic sequestration. In addition, Subpart W final rule
requires the reporting of any CO2 recovered at AGR units and transferred offsite. Between
todays final rule and the proposed subpart RR, EPA will be able to develop a mass balance of
all CO2 in an EOR system. Exempting AGR units from EOR operations will result in negative
emissions (or inflated sink) in the overall mass balance. Hence, EPA has retained the
requirement for all AGR units, including those in EOR operations, in todays final rule.
98.233(d) Acid Gas removal (AGR) vent stacks
In certain cases, calculating CO2 vent emissions via mass balance using the acid gas removal
unit inlet flow is more difficult and less accurate than by simply calculating the same emissions
using the flow to the CO2 vent and the CO2 mol% from a vent stream gas chromatography (GC)
analysis. The difficulty comes from the fact that elemental sulfur interferes with all methods of
inlet flow measurement. As a result of this interference, there are no flow meters on the acid gas
removal inlet. In order to calculate CO2 emissions, the acid gas removal unit inlet flow would
have to be back calculated via mass balance using the sales gas flow from the acid gas removal
unit outlet before it could be used to calculate CO2 emissions.
Since there are flow meters on each CO2 vent, and a vent stream sample is analyzed by GC, CO2
emissions could be calculated more efficiently and accurately by using the flow through the CO2
vent.
Therefore, the rule should include the option for facilities to calculate emissions using direct
measurement where the flow and composition of the gas being vented are being monitored
directly in the vent stack.
At certain other gas processing facilities equipped with cryogenic units to separate the ethane,
propane and butane (EPB) streams from the inlet gas, CO2 also separates with the EPB stream.
The CO2 rich EPB stream is then processed through an Amine Unit where CO2 is removed and
vented to the atmosphere. For such a facility the rule should include an option to use a mass
balance / engineering estimation approach to monitor / calculate CO2 vented to the atmosphere.
Such an alternative methodology should be detailed in the monitoring plan.
1479
The commenter did not provide adequate details on the EFB scenario for EPA to provide any
mass balance options. EPA reckons that if the EFB stream is eventually going through an AGR
unit then one of the four options provided in todays final should be applicable.

98.233 (d) The acid gas removal rule requires that you have a meter on the inlet and on the outlet
of your amine unit. Will it be permissible to use the values from more than one meter? (For
example, multiple wells with meters may flow into a compressor before flowing into the amine
unit. The natural gas is metered at the well. Can these well meters be summed to determine the
amount of gas flowing into the amine unit
Response: In todays final rule, EPA has provided multiple options to conduct engineering
estimation of flow rates. With regard to acceptable methods for calculating GHG emissions from
AGR units, please see the response to EPA-HQ-OAR-2009-0923-1024-26.
98.233 (d) Usually, glycol units are located downstream of the amine unit and the sales meter is
downstream of the glycol unit. Can the sales meter be used as the meter out of the AGR unit if
the water vapor removed by the glycol unit is subtracted from the total.
Response: In todays final rule, EPA has included additional options to estimate emissions from
AGR units. With regard to acceptable methods for calculating GHG emissions from AGR units,
98.233 Calculating GHG emissions.
Paragraph (d)(1) under Acid gas removal (AGR) vent stacks outlines the methodologies
acceptable to determine CO2e gas volumes. One method allows the use of pre and post AGR gas
1480
unit samples to determine the CO2e vented during operations. Current language states quarterly
samples must be taken to determine CO2e content. We request the sampling frequency be
reduced to every other year or annually as gas composition in these processes does not fluctuate
frequently enough to warrant quarterly sampling.
Response: EPA disagrees with the commenter. The commenter has not provided sufficient
details or analysis in the comment to support the claim that gas composition does not fluctuate
frequently. Furthermore, the acceptable levels of fluctuation are subjective and cannot be
commented upon without quantitative information. Finally, AGR units often receive natural gas
from several producing fields that feed varying volumes of natural gas throughout the year. This
leads to a constant fluctuation in the mixture of natural gas from these various feed sources.
Hence todays rule has retained the requirement for quarterly sampling of gas. However, EPA
has provided relief in terms of sampling on inlet stream only and allowed the use of pipeline
quality specification at the outlet stream. This will directly cut the sampling cost into half.
Commenter:
Source Type:
Acid Gas Removal Vent Stacks
40 CFR 98.233(d)
1. Metered total annual volume of natural gas flow into AGR unit in cubic feet per year at
ambient condition.
2. Volume weighted CO2 content of natural gas into the AGR unit.
3. Metered total annual volume of natural gas flow out of the AGR unit in cubic feet per year at
ambient condition.
4. Volume weighted CO2 content of natural gas out of the AGR unit.
Comment:
Acid gas emissions are calculated, not directly measured, using a mass-balance approach: Inlet
gas is measured as is outlet gas the difference is considered to be acid gas. This methodology
has been an accepted method with regulatory agencies. Yates gas is high in H2S, which is
highly corrosive: flow meters for acid gas have a high rate of failure and are highly unreliable
due to the highly corrosive environment. Therefore, Yates proposes an additional monitoring
methodology using the mass-balance approach to calculating acid gas.
1481

The rule should allow engineering estimation methods for acid gas removal
vent stack emissions. Proposed 98.233(d) prescribes the use of a mass balance equation for
calculation of emissions from acid gas removal vent stacks. In order to complete that equation,
meters must be located both upstream and downstream of the amine treater. Few operators have
such meters in place. Rather than prescribing use of mass balance equation, which would require
unnecessary expenditures in order to install additional meters, EPA should allow engineering
estimation (such as HySYS or Promax) to be used in place of mass balance equations

Section 98.233(d) Acid gas removal vent stack.
API supports EPAs selection of engineering methods to determine CO2 emissions from Acid
Gas Removal Vents. However, API has several suggestions to improve this methodology, lower
the burden imposed, and provide for rational implementation.
The requirements for Section 98.233(d) specifies metered total annual volume of natural gas
into the AGR unit and metered total annual volume of natural gas out of the AGR unit. The
use of the term metered implies the flow must be determined using a flow meter. Paragraph
98.233(d) does not reference Section 98.234(b) (requirements for flow meters, composition
analyzers and pressure gauges) and Table W-4 in the preamble specifies the monitoring method
for acid gas removal vent stacks is engineering estimate. Since Section 98.233(d) does not refer
to the flow meters requirements in Section 98.234(b) and the preamble does not indicate direct
measurement is required for acid gas vent stacks, it appears EPA intended flow to be based on
engineering estimate. Thus, API requests that the term metered be removed from Section
98.234(b) in regards to the volume of natural gas.
Requiring natural gas volume at both the inlet and outlet points is unnecessary, as the differential
in flow between the two points is almost 100% attributable to the change in acid gas content of
the gas entering and exiting the acid gas removal unit and CO2 emissions can be accurately
determined with quantifying gas flow at either the inlet or outlet of the acid gas removal unit
coupled, with sampling and analysis of the gas entering and exiting the unit.
1482
In addition, CO2 emissions can be accurately quantified by measuring the vent gas flow rate.
Due to elemental sulfur interference, there may not be a flow meter on the inlet of the AGR unit.
The inlet flow must be back calculated via mass balance using the sales gas flow from the outlet
from each plant before it can be used to calculate CO2 emissions. In some instances, a simpler
and more accurate way to quantify the CO2 emissions would be to use the flow to each CO2 vent
stream. The rule should include an alternative monitoring methodology for facilities where the
flow and composition of the gas vented though the AGR vent stack is being monitored directly in
the vent stack.
API requests that EPA remove the requirement for metering of gas at both the inlet and outlet
from the acid gas removal unit and enable metering or engineering estimate at one point (the
inlet, the outlet or the vent) to determine CO2 emissions from AGR units.
In addition, API requests the following:
- Acid gas removal units are highly efficient at removing CO2. As a result, it is not necessary to
measure the CO2 concentration of the outlet gas stream. It is a reasonable simplification to
assume that all of the CO2 in the inlet gas stream is removed.
- For acid gas removal units which do not have metering currently installed or where upgrades to
the current metering would be required to meet the rule accuracy requirements and which would
require a unit shut down to install or upgrade the required metering, API requests that EPA allow
the use of Best Available Monitoring Methods (BAMM) information until the next scheduled
shut-down of the unit where the metering could be installed or upgraded.
- For facilities with multiple acid gas removal units which are fed from a common pipe or header
arrangement, API requests that EPA enable a common pipe approach where the volume (entering
or exiting) and CO2 concentration is determined at the common pipe/header rather than at each
individual acid gas removal unit.
- The rule should include an alternative monitoring methodology for facilities where the flow and
composition of the gas vented though the AGR vent stack is being monitored directly in the vent
stack.
- The rule should provide a provision for acid gas removal units where the CO2 removed from
the inlet gas is captured and not vented to the atmosphere.- Acid gas removal vent stack
emissions is defined in Section 98.6 as acid gas separated from the acid gas absorbing
mediumand released with methane and other light hydrocarbons to the atmosphere or a flare.
Since acid gas removal vent stack emissions include gases routed to a flare, Section 98.233(d)
should be revised to clarify the calculation methodology in Section 98.233(d) applies when the
emissions are released to the atmosphere and the calculation methodology in Section 98.233(n)
applies when the emissions are routed to a flare.
Response: EPA agrees with the commenter on the use of engineering estimation of gas flow
rates to the AGR unit, including flow rate estimation for AGR units using a common inlet
header, under certain conditions; please see the response to EPA-HQ-OAR-2009-0923-1024-26.
It must be noted, however, that the flow rate and hence emissions have to be estimated and
reported for each AGR unit separately.
1483
EPA agrees with the commenter on the measurement of flow rates at the inlet and outlet streams
of an AGR unit. Todays final rule allows for the use of either the inlet or outlet stream flow rate
in the calculation. EPA also agrees with and allows the option of using vent stack meters to
estimate emissions from AGR units, including monitoring of a common vent stack connected to
multiple AGR units. It must be noted, however, that the emissions have to be estimated and
reported for each AGR unit separately.
EPA agrees with the commenter that AGR units are efficient in removing CO2. However, not all
of the CO2 is removed from the natural gas; pipeline quality gas contains some amount of CO2.
Hence, EPA has provided the option of using pipeline quality specification at the outlet stream so
as not to account for emissions that will not occur. Given the multiple options provided in
todays final rule, EPA does not consider it necessary to provide BAMM for this source
specifically. However, under certain conditions, the reporter may apply for a BAMM. Please see
Section II.F. of the preamble to todays final rule for further details.
With regard to AGR units where all or a portion of the CO2 removed from the inlet gas is
captured and not vented to the atmosphere, please see the response to EPA-HQ-OAR-2009
0923-1024-26.
EPA disagrees with the commenter on reporting of AGR emissions being sent to a flare as flare
emissions. For further clarification, please see the response to EPA-HQ-OAR-2009-0923-1305
25. However, in todays final rule EPA has made provisions to account for acid gas being sent to
a flare to be reported as AGR unit emissions.
Section 98.233(d) ) Acid gas removal (AGR) vent stacks. Paragraph 98.233(d) defines the
parameters V1 and V2 in Equation W-4 as "metered total annual volume of natural gas flow into
AGR unit" and "metered total annual volume of natural gas flow out of the AGR unit",
respectively. The use of the term "metered" implies the flow must be determined using a flow
meter. Paragraph 98.233(d) does not reference Section 98.234(b) (requirements for flow meters,
composition analyzers and pressure gauges) and Table W-4 in the preamble specifies the
monitoring method for acid gas removal vent stacks is engineering estimate. API requests
clarification that the volume of gas through the AGR unit can be based on engineering
estimation. In addition, the requirement to measure the flow rate out of the AGR unit (as shown
for Equation W-4) is not necessary as the difference in flow rates between the inlet and outlet are
minimal. This is consistent with the required emission estimation method for AGR units in the
California mandatory reporting regulation. Additional details on APIs suggested alternative
methods for acid gas removal are provided in Section VII.D of this document.
Response: Please see responses to EPA-HQ-OAR-2009-0923-1151-116 and EPA-HQ-OAR
2009-0923-1024-26 for a response to this comment.
1484

Acid Gas Removal Vents
We support EPAs proposed emission estimating method for acid gas removal vents that uses a
mass balance calculation measuring carbon content and gas volumes. Both CO2 and CH4
emissions should be estimated and reported. Hydrocarbons (such as methane) contained in the
incoming natural gas, may be absorbed in the amine stripping solution, and may be carried
through the process and subsequently emitted to the atmosphere or sent to a flare and oxidized to
CO2. We agree with WCI that gas sampling frequency should be increased to monthly, if
continuous gas analyzers are not installed.
Response: EPA disagrees with the comment on increasing gas sampling frequency from
quarterly sampling to monthly sampling. For further clarifications, please see the response to
EPA-HQ-OAR-2009-0923-0582-31.
GPA further recommends that ProMax and comparable programs be identified as acceptable
methods for calculating GHG emissions from AGR units. In addition, GPA recommends that
other mass balance calculations be allowed for AGR units. For example, an operator typically
has metered volumes and composition analysis of gas entering an AGR unit and can
conservatively assume 100% of the CO2 is removed (in some cases a few percent of the CO2
may not be removed). A simple calculation will provide accurate CO2 emissions from the AGR
unit without the need for downstream metering or analysis. In other cases, operators may have
meters and analysis on the AGR vent line and can provide a direct calculation of the CO2
emissions, again without the need for downstream metering or analysis. Subpart W should
provide flexibility for any of these methods to estimate the GHG emissions from AGR vents.
emissions from AGR units Please see the responses to EPA-HQ-OAR-2009-0923-1024-26 and
This section for acid gas removal vents requires gas metering and gas analysis into and out of the
unit, which is not a routine practice in the industry today. The regulation instead should allow
1485
conservative calculation assuming all CO2 is removed from the unit inlet, which only requires
metering and gas analysis into the unit. In most cases this is a very close estimate and only a few
percent of the CO2 is not removed. It also should allow use of commercially available process
simulators such as ProMax or other similar process simulators. These are industry accepted
sophisticated process simulators used across the industry to design these units and estimate
emissions for permitting purposes. See TSD at 42.
emissions from AGR units. Please see the responses to EPA-HQ-OAR-2009-0923-1024-26 and

Section 98.233(d) Acid gas removal (AGR) vent stacks
Paragraph 98.233(d) defines the parameters V1 and V2 in Equation W-4 as "metered total annual
volume of natural gas flow into AGR unit" and "metered total annual volume of natural gas flow
out of the AGR unit", respectively. The requirement to measure the flow rate both in and out of
the AGR unit (as shown for Equation W-4) is not necessary as the difference in flow rates
between the inlet and outlet are minimal. This is consistent with the required emission estimation
method for AGR units in the California mandatory reporting regulation.
Response: EPA agrees with the comment; please see the response to EPA-HQ-OAR-2009-0923
1024-26 for further details. .
Section 98.233(d) Acid gas removal vent stack.
BP supports EPAs selection of engineering methods to determine CO2 emissions from Acid
Gas Removal Vents. However, BP has several suggestions to improve this methodology, lower
the burden imposed, and provide for more streamlined implementation.
1. The current rule requires metering of gas flow into and out of each acid gas removal unit
(mostly amine/solvent contactor vessels). Requiring metering at both points is unnecessary as the
differential in flow between the two points is almost 100% attributable to the change in acid gas
content of the gas entering and exiting the acid gas removal unit and CO2 emissions can be
accurately determined with metering of gas flow at either the inlet or outlet of the acid gas
removal unit coupled with sampling and analysis of the gas entering and exiting the unit. BP
1486
requests that EPA remove the requirement for metering of gas at both the inlet and outlet from
the acid gas removal unit and enable metering at either point. The methodology for determining
emissions (98.233 (d)) would be modified as follows:
(d) Acid gas removal (AGR) vent stacks. For AGR (including but not limited to processes such
as amine, membrane, molecular sieve or other absorbents and adsorbents), calculate emissions
for CO2 only (not CH4) using Equation W-4 of this section. For Volume measurement entering
the AGR: Ea,CO2 = (V*(Vol1% - Vol2%)) * 52.605 metric tonnes CO2 per MMSCF (Eq. W-4a)
For Volume measurement exiting the AGR:
Ea,CO2 = ((V+(V*(Vol1% - Vol2%) * (Vol1%-Vol2%))*52.605 metric tonnes CO2 per
MMSCF (Eq. W-4b)
Where:
- Ea,CO2 = Annual mass CO2 emissions in metric tons (volume is determined at STP conditions
of 60 degrees f and one atmosphere pressure (14.7 psi), in MM standard cubic feet per year)
- V = Metered total annual volume of natural gas flow entering or exiting the AGR unit in
standard cubic feet per year at 60 degrees F and one atmosphere condition.
- %Vol1 = Volume weighted CO2 content of natural gas into the AGR unit.
- %Vol2 = Volume weighted CO2 content of natural gas out of the AGR unit.
- 52.605 metric tonnes of CO2 per million standard cubic feet (60 degrees f and one
atmosphere pressure) of CO2
(1) If a continuous gas analyzer is installed, then the continuous gas analyzer results must be
used. If continuous gas analyzer is not available, quarterly gas samples must be taken to
determine %Vol1 and %Vol2 according to methods set forth in Section 98.234(b).
(2) Mass CO2 emissions shall be calculated from volumetric CO2 emissions using calculations
in paragraphs (u) and (v) of this section.
Example:
Metered gas flow - inlet: 38,321.35 MMSCF
Metered gas flow outlet: 36,500 MMSCF
Vol%1 CO2: 5.00% molar/volume
Vol%2 CO2: 0.10% molar/volume
Measurement at AGR inlet: (38,321.35 MMSCF*(5.00%-0.10%))*52.605 metric tons per
MMSCF = 100,593 metric tons CO2
1487
Measurement at AGR outlet:

((1000 MMSCF + (1000 MMSCF*(5.00%-0.01%)) * (5.00%- 0.01%)*52.605 metric tons per
MMSCF = 100,593 metric tons CO2
*Note: Both metered gas flow rates are equivalent to a 100 MMSCF per day outlet flow rate for
365 days
a. For acid gas removal units which do not have metering currently installed or where upgrades
to the current metering would be required to meet the rule accuracy requirements and which
would require a unit shut down to install or upgrade the required metering, BP requests that EPA
allow the use of Best Available Monitoring (BAM) information until the next scheduled shut
down of the unit where the metering could be installed or upgraded.
b. For facilities with multiple acid gas removal units which are fed from a common pipe or
header arrangement BP requests that EPA enable a common pipe approach where the volume
(entering or exiting) and CO2 volume % differential is metered and determined at the common
pipe/header rather than at each individual acid gas removal unit.
- The rule should include an alternative monitoring methodology for facilities where the flow and
composition of the gas vented though the AGR vent stack is being monitored directly in the vent
stack.
- The rule should provide a provision for acid gas removal units where the CO2 removed from
the inlet gas is captured and not vented to the atmosphere.
Response: EPA agrees with the commenter on the measurement of flow rates both at the inlet
and outlet streams of an AGR unit. Todays final rule allows for the use of either the inlet or
outlet stream flow rate in the calculation; please see response to EPA-HQ-OAR-2009-0923
1024-26.
In regards to capture of CO2 at AGR units, please see response to EPA-HQ-OAR-2009-0923
1024-26.
In regards to the monitoring of vent stacks, use of common headers for feed into AGR, and
BAMM please see response to EPA-HQ-OAR-2009-0923-1151-116.
13.2.5
FLARE CALCULATIONS

98.233 (f):
1488
With respect to Equation W-6 CAPP recommends that this equation should be a summation of
all individual events since the flow rate will change and facilities would have to sum all events to
be able to accurately determine an "average" flow rate.
Under 98.233(f)(1)(i) CAPP wanted to bring it to the attention of the EPA that the measured
flow rate of gas is likely to be non-linear with time as initially the production tubing contains
significant water which is displaced over time. The amount of water the tubing contains will
affect the volume of gas released in at a given time.
CAPP noted that equation W-7 incorrectly refers to casing diameter, when it should be
referencing tubing diameter. Additionally, CAPP was unable to locate the source of Equation W
7 in 98.233(f)(1)(ii) and therefore we were unable to determine the basis for using a combination
of "the volume of the production tubing down to the well depth" plus "sales rate times venting
hours". In addition CAPP believes the use of sales rate in equation W-7 is not valid since it will
fluctuate with changes in back-pressure. For this reason CAPP recommends the use of a simple
choked-flow (tubing area) orifice equation like in Section (g)(ii) since down hole reservoir
pressure is much greater than two times atmospheric pressure. (CAPP, 2002. Estimation of
Flaring and Venting Volumes from Upstream Oil and Gas Facilities. Section 3.2.1).
If the EPA wishes to continue with the existing published equation W-7, CAPP recommends
referencing the production tubing diameter and not the casing diameter.
Response: EPA disagrees with the commenter. Todays final rule includes an equation where
liquids unloading to the atmosphere is done with and without the aid of a plunger lift. When a
well is unloaded to the atmosphere without a plunger lift, more than the tubing volume of the gas
goes to the atmosphere as explained subsequently. In developing the emissions estimating
options for well liquids unloading, EPA considered the options reported to the EPA Natural Gas
STAR by companies who directly measured these emissions before and after installing plunger
lift systems. EPA collaborated with Partners to develop the equation in the Natural Gas STAR
Lessons Learned study Installing Plunger Lift Systems in Gas Wells,
http://www.epa.gov/gasstar/documents/ll_plungerlift.pdf, Appendix: Alternate technique for
calculating avoided emissions when replacing blowdowns. The logic of this alternate technique,
which reasonably matches Partner reported measurements, is that when liquid accumulation in
the gas well tubing reaches a head which, combined with the sales line back pressure, suppresses
gas production to near zero, it is equivalent to gradually shutting in the well. As this occurs, the
reservoir pressure surrounding the well perforations rises to shut-in pressure, which accumulates
in the well casing. If well gas production is stopped altogether by liquid accumulation in the
tubing, it is the same as shutting in the well at the surface: the liquid will slump to the bottom of
the tubing and the casing will be pressured up to shut-in pressure. This gas bubble blows the
liquid accumulated in the tubing to the surface during a well unloading. It is correct that the gas
in the tubing above the liquid, at sales line pressure, would be expelled first, but this is a small
fraction of the gas bubble in the casing at shut-in pressure that pushes the liquid up the tubing to
the surface. Natural Gas STAR Partners have reported that blowing liquids without the aid of a
plunger lift is very inefficient as liquid coats the inner wall of the tubing and runs back down the
well. Hence EPA assumed that all of the casing gas is required to clear significant portion of the
1489
liquids and provided for operators continuing to blow the well beyond the estimated time to blow
off the casing gas pressure.
EPA has added the average sales rate times the venting hours to the equation. EPA understands
from Natural Gas STAR experience that operators may sometimes leave a well open to the
atmosphere longer than the time taken to blow-off the casing and tubing volumes, especially if it
takes a long time for the well unloading to conclude. In such cases, EPA intends to capture the
amount of gas lost between the well unloading completion and the time the well is opened back
to the sales line. EPA recognizes that in the proposed rule there was some overlap of emissions
in the equation provided. However, for todays final rule EPA has calculated how long it would
take to blow off the bubble of gas in the casing up the tubing string in the average well: one hour
(see Memo 6: Change to Rule Equation W-7:). Hence, EPA has adjusted the sales rate times
venting hours by a factor for continued blowing a well to the atmosphere at average sales
production rate for the period of time beyond one hour that the operator leaves the well open.
EPA has also provided a calculation for blowing wells to the atmosphere with the aid of a
plunger lift. This is necessary when the reservoir energy is no longer sufficient to lift both the
plunger and liquid load against the sales line back pressure. For this calculation, EPA has
determined that the volume of gas in the tubing at sales line pressure, with average sales flow
rate beyond the estimated half-hour to expel the gas and liquid in the tubing is a conservative
estimate. EPA recognizes that the well flow to sales fluctuates, and that this is an uncertain,
highly variable way to estimate excess flow to the atmosphere after initial blow-off of the gas in
the tubing and casing. However, considering the burden of installing a recording flow meter on
every well that is blown to the atmosphere, EPA determined that this is a sufficient means of
quantifying this excess venting to inform future policy.
EPA is not recommending a calculation across the well choke because this can be a complex
calculation with sonic flow part of the time and subsonic flow for all or the remainder of the
unloading time. The pressure upstream of the tubing string or choke is very uncertain with
liquids being lifted inefficiently without a plunger lift and highly variable depending on the
amount of liquid accumulation at the time a well is selected for liquid unloading. EPA has
decided that the direct measurement of vent gas or the equations based on well depth, casing
diameter, tubing diameter, shut-in and sales line pressures, and venting time will be costeffective and adequately accurate in characterizing emissions.
98.233 (j):
Under 98.233(j)(1) CAPP believes that collecting separator temperature and pressure is not
appropriate as the product will already be at atmospheric temperature and pressure due to
transportation conditions.
With respect to 98.233(j)(2)(i) CAPP believes the words "or to flare" should be incorporated.
Additionally there is a spelling error: benefitial should be "beneficial".
1490
Could the EPA provide clarity to CAPP on the reference for the 3.87 and 5.37 factors used in
98.233(j)(4)(ii) and (iii)?
Response: EPA disagrees with the commenter on collecting separator temperature and pressure.
The software programs that use Peng-Robinson equation require separator temperature pressure
as an input to estimate flashing losses at the tank. The amount of vapors retained in the
hydrocarbon liquids post separation phase is directly correlated with the temperature and
pressure at the low pressure separator. Lower volumes of vapors are separated at low
temperatures and high pressures. These vapors may then attain ambient conditions downstream,
but they are nonetheless flashed through the tank. Hence using temperature and pressure from
the low pressure separator is vital to an accurate representation of flash emissions.
EPA agrees with the commenter on the monitoring requirement to provide for tank emissions
going to a flare and has made corrections to the typographical error. Please refer to Appendix E
of the TSD for clarification and discussion for the 3.87 and 5.37 factors used in 98.233(j)(4)(ii)
and (iii).
In 98.233(l) (3)(ii)(1)(1) CAPP would like clarification on how the GOR is to be determined.
Will it be based on a 1-hour test, a 24-hour test or some other time period or estimation method?
CAPP notes that the requirement in Alberta is a 24 h test that is to be updated annually if GOR
>100 m3/m3 and every three years if GOR < 100 m3/m3. In addition CAPP would like clarity
from the EPA on whether there is a list of approved "consensus-based standards organizations".
Response: The intent of todays rule is to gather relevant information with reasonable burden to
industry. Hence, in todays final rule, GOR can be determined by two methods; (1) using
available data representative of the companys facility, and (2) using standard consensus-based
methods or standard industry practice. EPA has determined that using any consensus based
standard will provide reasonable data quality to inform policy. Hence, in todays final rule, EPA
is not providing a list of approved standards, so as to not constrain the reporters, and allow them
the flexibility of choosing the most appropriate standard. Some examples of approved consensusbased standards organizations are American Society for Testing and Materials, American
National Standards Institute, American Petroleum Institute Standards, and American Gas
Association.
98.233 (m):
1491
Again for 98.233(m)(1)(i), CAPP would like clarity from the EPA on whether there is a list of
approved "consensus-based standards organizations".
CAPP would also appreciate comment from the EPA on whether or not it is necessary to
determine the GOR for each well. In cases where the associated gas is flared, and the gas from
the separator and treater is often metered before it is flared. This would give a direct
measurement of the volume of gas flared.
Response: In todays final rule, EPA allows for use of GOR as determined for a cluster of wells
within the same field to be used if individual well GOR is not available. With regards to a list of
consensus-based standards organizations, please see the response to EPA-HQ-OAR-2009-0923
1018-33.
CAPP would like to draw EPAs attention to equation W-14 in 98.233(n)(4), this equation is
correct only if Yj includes carbon dioxide as well as the hydrocarbons listed.
Response: EPA disagrees with the commenter. The equation W-14 that estimated combustion
emissions using Yj only accounts for hydrocarbons that are converted into CO2, not the trace
CO2 in product streams. Hence the equation has been retained. However, EPA did correct the
other flare stacks equations (W-13 and W-15 in the supplementary proposed rule) to account for
the un-combusted CO2 in the gas stream being sent to the flare. Please see the response to EPA
HQ-OAR-2009-0923-0582-29 and EPA-HQ-OAR-2009-0923-1074-36.

98.233(l) Well Testing Venting and Flaring:
a.) The proposed rule assumes that all emissions from well testing are either vented to the
atmosphere or flared. In our Alaska operations, all gas generated from routine well testing is
routed to the fuel gas system for use as fuel or for reinjection into the reservoir. Therefore, we
request that the calculation methodology allow for beneficial use and be excluded from reporting
emissions. See ConocoPhillips General comment for GHGs to report.
b.) This paragraph references the use of 98.233(u) for calculating GHG emissions. Paragraph
(u) specifies quarterly sampling something infeasible for the manner in which our Alaska
1492
facilities will be subject to this category; i.e., in the very infrequent and non-routine drill stem
tests on exploration and other new wells where we use a portable flare and work to understand
well characteristics in advance of production. Quarterly sampling for these would be costly and
would produce no useful data. Thus, we request that EPA allow for a single, rather than a
repeated quarterly, gas sample for each new reservoir.
Response: For the well testing venting and flaring source, the source monitoring method
description clarifies that the emissions are to be reported only when there is venting or flaring.
Hence, the commenters concern with regards to modifying the calculation methodology to allow
for beneficial use and being excluded from reporting is without any basis.
After considering several comments, EPA simplified the requirement for calculating GHG
emissions by allowing for the use of existing composition data. Hence, this should mitigate any
technical or cost constraints.

Commenter:
Source Type:
Well testing venting and flaring
40 CFR 98.233(l)
1. Gas to oil ratio in cubic feet of gas per barrel of oil; oil here refers to hydrocarbon liquids
produced of all API gravities.
2. Flow rate in barrels of oil per day for the well being tested.
3. Number of days during the year, the well is tested.
Comment:
It is not industry standard to determine GOR by individual wellhead. Yates requests the use of
Response: In todays final rule, EPA allows for use of GOR as determined for a cluster of wells
within the same field to be used if individual well GOR is not available. This allows flexibility
for the reporter and manages burden. EPA still recognizes that the gathering of data from wells
soon after the promulgation of this rule might pose a challenge to some reporters and therefore
EPA is allowing the use of best available monitoring methods from this source from January 1,
2011 until June 30, 2011. For further details on best available monitoring methods for emissions
from well venting and flaring, please see the Section II.F of the preamble to todays final rule.
1493

GPA recommends that HYSYS and other comparable process simulators be identified as
acceptable methods for calculating GHG emissions from tanks, in addition to E&P Tanks. The
use of HYSYS or comparable process simulators for tank calculations is particularly important
due to recognized difficulty in collecting acceptable separator liquid samples when using E&P
Tanks. In some case, the success rate for collecting acceptable separator liquid samples can be as
low as 10%. HYSYS and similar programs provide the flexibility to use gas analysis to model
tank emission.
Response: EPA agrees with this comment. Hence, in todays final rule EPA allows the use of
any simulation software that conforms to the requirements in Section 98.233(j) of todays final
rule.
This requirement states that an operator must report the proportion of total natural gas to pure
hydrocarbon stream being sent to a flare annually for the reporting period. This value is not
required by the calculation methodology, so this requirement only creates an unnecessary
recordkeeping and reporting burden on the operator. We request that EPA remove this
requirement.
Response: EPA agrees with the comment that reporting the proportion of total natural gas to
pure hydrocarbon stream being sent to flare annually creates unnecessary recordkeeping and
reporting burden since the data is not readily available with the operator. Hence, EPA has
removed this requirement from Section 98.236 of todays final rule.

Section 98.233(l) Well testing venting and flaring
Section 98.233(l) requires determining the gas to oil ratio (GOR) of the hydrocarbon production
for each well tested, determining the flow rate of oil for the well being tested, and tracking the
number of days in the year that the well is tested.
The vast majority of well tests are conducted while the wells are in operation and do not require
1494
venting or flaring. As a result, the rule should exempt well testing that does not result in venting
or flaring natural gas.
Response: The commenters concern is without any basis. With regards to well testing that does
not result in venting or flaring, please see the response to EPA-HQ-OAR-2009-0923-1042-20.
13.2.5.1
FLARE STACKS

WCI recommends using the USEPA Subpart W methodology in SECTION 98.233(n).
WCI Modifications: WCI has identified a problem with the current USEPA methodology found
in SECTION 98.233(n). We are currently seeking clarification from USEPA concerning this
issue which is discussed below.
Calculation of Flare Stack Emissions (SECTION 98.233(n)(4))
We assume that the intent of the rule is to account for the following emissions from flare stacks:
1) uncombusted methane
2) carbon dioxide produced by combustion in the flare
3) carbon dioxide present in the inlet gas to the flare and passing through it
This section and Equations W-13, W-14, W-15 therein appear to be incorrect and confusing. For
example:
- Calculation of CO2 emitted as pass-through from the inlet gas would presumably be accounted
for by Equation W-13, but for CO2 the value of n (percent combusted) would be zero, not 98%.
- Equation W-15 includes the term E [subscript a, i] (combusted), but the equation as given is
incorrect when i = CH4. Since the TSD characterizes combustion methane emissions as
negligible, we assume the intent is to ignore these negligible emissions and calculate only the
uncombusted methane emitted from the flare stack.
We recommend a more explicit calculation of the emissions as follows:
(4) Calculate GHG volumetric emissions at actual conditions using Equations W-13, W-14, and
1495
E [subscript a, CH4] = V [subscript a] * (1 ?n ) * X [subscript CH4] (Eq. W-13)

E [subscript a, CO2] (non ? combustion ) = V [subscript a] * X [subscript CO2] (Eq. W-14)
E [subscript a, CO2] (combustion) = [sigma symbol with j below] (n * V [subscript a] * Y
[subscript j] * R [subscript j] (Eq. W-15)
E [subscript a, CO2] (total) = E [subscript a, CO2] (non ? combustion) + E [subscript a, CO2]
(combustion) (Eq. W-16)
Where:
E [subscript a, CH4] = Contribution of annual uncombusted CH4 emissions from flare stack in
cubic feet, under ambient conditions.
E [subscript a, CO2] (non-combustion) = Contribution of annual CO2 emissions from CO2 in
inlet gas passing through flare stack.
E [subscript a, CO2] (combustion) = Contribution of annual emissions from combustion from
flare stack in cubic feet, under ambient conditions
E [subscript a, CO2] (total) = Total annual CO2 emissions from flare stack in cubic feet, under
ambient conditions
V [subscript a] = Volume of natural gas sent to flare in cubic feet, during the year.
n = Percent of natural gas combusted by flare (default is 98 percent).
X [subscript i] = Concentration of GHG i in gas to the flare.
Y [subscript j] = Concentration of natural gas hydrocarbon constituents j (such as methane,
ethane, propane, butane, and pentanes plus).
R [subscript j] = Number of carbon atoms in the natural gas hydrocarbon constituent j; 1 for
methane, 2 for ethane, 3 for propane, 4 for butane, and 5 for pentanes plus).
Response: EPA agrees with the comment and has sufficiently modified the equations related to
flare stacks emission calculation to address all of the issues. The equations now capture un
combusted CH4 emissions and total CO2 emissions, where total CO2 emissions include un
combusted and combusted CO2 emissions.

When flares are used in emergency situations, concentration of GHG in the gas flared is not
immediately available. PAW requests clarification that it is acceptable to calculate GHG
emissions during upset conditions using available concentration data, and correcting the
concentration as more accurate data becomes available following an upset condition.
1496
Response: EPA agrees with the use of available concentration data and subsequent correction
during emergency conditions. Please see response to EPA-HQ-OAR-2008-0508-0952-1, excerpt

Commenter:
Source Type:
Flares
40 CFR 98.233(n)
1. Volume of natural gas sent to flare in cubic feet, during the year.
2. Percent of natural gas combusted by flare (default is 98 percent).
3. Concentration of GHG i in gas to the flare.
4. Concentration of natural gas hydrocarbon constituents j (such as methane, ethane, propane,
butane, and pentanes plus).
5. Number of carbon atoms in the natural gas hydrocarbon constituent j; 1 for methane, 2 for
ethane, 3 for propane, 4 for butane, and 5 for pentanes plus).
Comment:
When flares are used in emergency situations, concentration of GHG in the gas flared is not
immediately available. Yates requests clarification that it is acceptable to calculate GHG
emissions during upset conditions using available concentration data, and correcting the
concentration as more accurate data becomes available following an upset condition.
Response: In todays final rule, the calculation method for flare stacks requires the use of
representative composition of gas stream being sent to the flare. In regards to use of available
concentration data, please see response EPA-HQ-OAR-2009-0923-1015-12 for further details.
Commenter:
Comment on 98.233(n):
WBIH requests clarification on the "un-combusted" emissions. Is this considered to be 2% when
using 98% flare combustion efficiency?
1497
Response: EPA has clarified the equations for estimating emissions from flare stacks in todays
final rule. The un-combusted emissions are evaluated in two portions; un-combusted CO2 and
un-combusted CH4 emissions. Un-combusted CO2 emissions refers to the entire CO2 volume
present in gas being sent to flare. Un-combusted CH4 emissions refers to the CH4 present in the
un-combusted portion of the flare gas; if flare efficiency is X then un-combusted CH4 emissions
refers to the volume of CH4 present in (1-X) percent volume of the flare gas.
Organization1: American Petroleum Institute
25. Section 98.233(n)(2)(i) Flare stacks. The terminology utilized in this section is not
representative of many gas plants, which do not utilize de-methanizers. De-methanizers are
typically utilized in natural gas liquid recovery or fractionation plants. A more appropriate name
for use in place of De-methanizer would be hydrocarbon dewpoint control plant. What gas
composition is applied if a plant does not utilize a De-methanizer?
Response: In regards to calculations for gas processing facilities without de-methanizers, todays
final rule has been updated to require dewpoint control point as the demarcation point. The dew
point control is equivalent to a de-methanizer in terms of compartmentalizing the processing
facility into two major streams of gas compositions, hence this change should provide sufficient
clarity on what gas composition to use for estimating emissions. For composition of gas, please
see Section 98.233(n) of todays final rule.

26. Section 98.233(n)(4) Flare stacks. Flare emission equations W-13, W-14, and W-15. As
currently presented, Eq. W-13 assumes 2% of the CO2 present in the flared gas is uncombusted.
CO2 already present in the gas stream is not impacted by flare efficiency; all of the CO2 present
in the flared stream is emitted as CO2. We suggest the following equations for CO2 and CH4
emissions from flares:
Eq. a: E [subscript CO2, a, i] = [sigma symbol with N on top and 1 on bottom] (V [subscript a] x
X [subscript CO2] + n x V [subscript a] x [sigma symbol with i on top and 1 on bottom] (Y
[subscript j, i] x R [subscript j]) )
Eq. b: E [subscript CH4, a, i] = [sigma symbol with N on top and 1 on bottom] (V [subscript a] x
(1-n) x X [subscript CH4])
Where:
1498
E [subscript CO2, a, i] = Annual CO2 volumetric emissions from flare stack at actual conditions
for gas analysis i.
E [subscript CH4, a, i] = Annual CH4 volumetric emissions from flare stack at actual conditions
for gas analysis i.
V [subscript a] = Volume of natural gas sent to flare stack at actual conditions determined from
Section 98.234(j)(1).
H = Percent of natural gas combusted by flare (default is 95 percent for non- steam aspirated
flares and 98 percent for steam aspirated or air injected flares).
X [subscript CO2] = Molar concentration of CO2 in the flare gas determined from Section
98.234(j)(1).
X [subscript CH4] = Molar concentration of CH4 in the flare gas determined from Section
98.234(j)(1).
Y [subscript j, i] = Concentration of natural gas hydrocarbon constituents j (such as CH4, ethane,
propane, butane, and pentanes plus) for gas analysis i.
R [subscript j] = Number of carbon atoms in the natural gas hydrocarbon constituent j; 1 for
methane, 2 for ethane, 3 for propane, 4 for butane, and 5 for pentanes plus).
N = Number of required gas analyses during reporting year.
In addition, Equation W-15 adds scf of CH4 emissions to scf of CO2 emissions. The alternative
equations provided above correct this problem.
Response: EPA has clarified flare stack emissions equations. For further details, please see the
response to EPA-HQ-OAR-2009-0923-1074-36 and EPA-HQ-OAR-2009-0923-0582-29.

27. Section 98.233(n)(8) Flare stacks. The phrase should read Any emissions calculated under
this source are excluded from other emission sources in Section 98.233.
Response: EPA disagrees with the comment. With regard to reporting requirements in Section
98.233(n)(8) of todays final rule, please see the response to EPA-HQ-OAR-2009-0923-1305-25.

This section provides flexibility to use engineering calculations, company records, or similar
estimates to determine volumetric flow to a flare. However, if a flare currently has a continuous
flow meter, the proposal required it must be used to determine flow to the flare. The experience
of GPAs membership is that existing flare meters may be accurate only at certain flow regimes
and not for all flows from all sources to the flare. If existing meters are used to the exclusion of
engineering estimates or other estimation methods, accuracy for flow determination to the flare
1499
may be reduced. GPA recommends that an operator be allowed to use any combination of
meters, engineering calculations, company records, or similar estimates to determine volumetric
flow to a flare. The operator can then determine what combination of metering and calculation
will provide the most accurate estimate of flow to a flare for the purposes of estimating GHG
emissions.
Response: After considering these comments, EPA has revised language in todays final rule to
provide for sufficient flexibility in using the continuous flow meters for ranges in which they are
accurate and use of other options provided in the rule otherwise. Please see response to section
98.233(n) of todays final rule for further details.

P. Section 98.233(n)(2)(i) Flare stacks
The terminology utilized in this section is not representative of many gas processing facilities,
which do not utilize de-methanizers. De-methanizers are typically utilized in natural gas
liquid recovery or fractionation plants. What gas composition is applied if a plant does not utilize
a De-methanizer?
Q. Section 98.233(n)(8) Flare stacks.
The phrase should read Any emissions calculated under this source are excluded from other
emission sources in 98.233.
Response: Regarding comments related to the de-methanizer, please see response to EPA-HQ
EPA agrees and has revised todays final rule in 98.233(n) to correct flare emissions to avoid
double counting.
13.2.5.2
ASSOCIATED GAS VENTING AND FLARING

WCI recommends adoption of the calculation methodology set-forth by US EPA in section
1500
SECTION 98.233(l) for vented and flared emissions resulting from well testing with minor
modifications.
WCI Modifications:
WCI recommends deleting the following language. Deletion of this text will insure that GOR is
determined by actual measurement and insure more accurate and consistent data.
SECTION 98.233(l)(1)(i) If GOR is not available then use an appropriate standard method
published by a consensus-based standards organization to determine GOR.
Response: EPA considered the commenters input however EPA has retained the requirements
for GOR determination from the April 2010 proposed rule found in docket (EPA-HQ-OAR
2009-0923-0002). EPAs intent is to gather relevant information with reasonable burden to
industry. In the case of well GOR, oil and gas reporters will typically know their GOR, since it is
normal practice to estimate GOR regularly. Hence use of this existing GOR value is deemed
sufficient to estimate emissions from this source type.

WCI recommends adoption of the method (SECTION 98.233(m)) published by US EPA in the
most recent Subpart W with minor modifications.
WCI Modifications:
WCI recommends deleting the following language. Deletion of this text will insure that GOR is
determined by actual measurement and insure more accurate and consistent data.
SECTION 98.233(l)(1)(i) If GOR is not available then use an appropriate standard method
published by a consensus-based standards organization to determine GOR.
Response: EPA disagrees with the comment. With regard to modifications in Section 98.233(l)
of todays final rule, please see the response to EPA-HQ-OAR-2009-0923-0582-27.

1501
K. Section 98.233(m) Associated gas venting and flaring

Onshore Production Source Category: Associated gas venting and flaring
Section 98.233(m) requires determining the GOR of the hydrocarbon production from each well
whose associated gas is vented or flared.
GOR determination for estimating emissions from venting and flaring events should be
simplified. API suggests that the GOR be determined using an appropriate standard industry
practice or method published by a consensus-based standards organization for a representative
sample of wells within the Sub-basin entity whose associated gas is vented or flared.
Response: EPA understands that the GOR from each well is periodically monitored by oil and
gas operators. Also, in todays final rule EPA has clarified that it is not requiring any new
measurements unless no data is available. Hence, EPA has determined that there is sufficient
flexibility available to reporters to estimate emissions with minimal burden. Therefore, EPA has
retained the requirements for GOR determination from the proposed rule for estimation of
emissions from individual wells.
13.2.6
BLOWDOWN VENT STACKS
Commenter:

The final example is related to blowdown vent stacks. There are a limited number of tasks in the
natural gas industry that result in blowing down significant volumes of natural gas to the
atmosphere. These activities are limited to blowing down compressors, process vessels and
piping. There are other tasks in the natural gas industry that result in blowing down very small
and insignificant volumes of natural gas, such as blowdowns from sight glasses and small tubing.
Without thresholds which limit the blowdowns that must be recorded, calculated, and reported,
this portion of the proposed Subpart W will be unmanageable. GPA recommends that GHGs
only from the following blowdowns be reported: 1) all compressor blowdowns and 2)
blowdowns from process vessels & piping with a physical volume of 50 cubic feet and larger.
Response: EPA does not agree with the commenter on limiting blowdowns to compressors only.
There are other vessels that when blown down result in large emissions, which EPA intends to
capture. However, EPA agrees with the commenter on providing an equipment threshold. EPA
conducted its own analysis which indicates that the 50 cubic feet equipment threshold is
reasonable. Please see Equipment Threshold for Blowdowns memo in rulemaking docket
EPA-HQ-OAR-2009-0923 for further details. Hence, in todays final rule, EPA has provided a
50 cubic feet equipment threshold. Reporters do not have to report blowdown emissions from
any blowdown volume below this equipment threshold.
1502

Blowdown Vent Stacks
Source Description: Natural gas emissions can occur whenever equipment such as a compressor
are blowndown that is the pressurized gas contained in the equipment is released to the
atmosphere, sent to a vent stack or control device such as a flare. Routine shutdowns for
maintenance and emergency shutdowns both result in release of natural gas.
Measurement Options: The approach recommended by USEPA was considered to be
implementable without undue burden and sufficiently accurate to support a cap and trade rule.
No other options were evaluated by WCI.
USEPA Approach: USEPA provides the following calculation methodology for blowdown stack
vent emissions.
Step 1: Calculate the total volume (including, but not limited to, pipelines, compressor case or
cylinders, manifolds, suction and discharge bottles and vessels. Variable V [subscript v] in
equation below.
Step 2: Record the number of blowdowns for each equipment type (N in equation below), and
Step 3: Calculate annual venting emissions using the following formula:
E [subscript a, n] = N * V [subscript v] (Eq. W-10)
Where:
E [subscript a, n] = annual natural gas venting emissions at ambient conditions from blowdowns
in cubic feet
N = number of blowdowns for the equipment in the reporting year
V [subscript v] = total volume of blowdown equipment chambers (including, but not limited to,
pipelines, compressors and vessels) between isolation valves, in cubic feet.
Step 4: Calculate natural gas volumetric emissions at STP using calculations in paragraph (t) of
this section
Step 5: Calculate CH4 and CO2 volumetric emission using methods in paragraph (u)
Step 6: Calculate CH4 and CO2 mass emissions using methods in paragraph (v) of this section.
WCI Recommendation: We recommend using the approach provided by USEPA without any
1503
modification.
Response: EPA has made modifications to the blowdown methodology to allow for the
estimation of blowdown volumes with differing temperature and pressure conditions. With
regard to modifications in equations for blowdown vent stacks, please see the response to EPA
HQ-OAR-2009-0923-1011-41.
XII. Proposed Rule Section 98.233(i) Blowdown vent stacks.El Paso requests that an alternate
method be added to allow blowdown volumes be corrected to standard conditions for each event
prior to aggregation.
Response: EPA agrees with the comment. With regard to modifications in equations for
blowdown vent stacks, please see the response to EPA-HQ-OAR-2009-0923-1011-41.

XII. Section 98.233(i) Blowdown vent stacks
The proposed regulation requires first aggregation of blowdown volumes at actual conditions
from all events at the facility and then the conversion of the total volume to standard conditions.
The regulation fails to take into account that each blowdown event may occur at a different
pipeline pressure and a different gas temperature. In practice, the blow-down volumes are
corrected to standard conditions for each event and then aggregated. Having the order of
calculations as proposed would result in having to use a pressure of MAOP (Maximum
Allowable Operating Pressure) and perhaps yearly average temperature at the site which would
probably result in an over predictions of the blowdown volumes. Therefore, El Paso recommends
that the reporters be allowed to correct the blowdown volumes to standard conditions on a per
event basis.
Further, the proposed methodology is based on an assumption that the total volume between the
isolation valves is blowdown. In practice, frequently a blowdown continues only until a desired
working pressure is achieved. The desired working pressure may differ for each blowdown event
depending on the reason for the blowdown. Therefore, treating each blowdown event
individually and considering the initial and ending pressures will result in more accurate
emission estimates.
It should be noted that many companies currently employ algorithms and programs to calculate
1504
and document blowdowns, and these methodologies use the same ideal gas law principles as
intended by the Subpart W equations to calculate event-based blowdown volumes. Companies
currently employing equivalent methods should be allowed to use their methodologies for the
purpose of reporting GHG emissions under Subpart W.
In addition, El Paso recommends an option be provided for the installation of permanent meters
on blowdown vent stacks.
Response: EPA agrees with the commenter with respect to different blowdowns being at
different operating pressures. EPA has revised todays final rule equation for blowdowns to
account for different blowdown volumes at different pressures.
In todays final rule, EPA clarifies that the blowdown volumes to be reported are for complete
blowdowns to atmospheric pressure when equipment is being taken out of service, and does not
include operating pressure adjustments. This means that any equipment and associated vessels
and pipelines that vent gas to the atmosphere to bring actual pressure back to operating pressure
do not report the resultant emissions. This clarification has been provided in todays final rule.
Given the clarifications and corrections, EPA determined that the calculation methods based on
equipment volume are uniform across many reporters, sufficiently accurate, and cost-effective to
estimate and report blowdown emissions. Hence, EPA has not allowed for any new methods or
for installation of meters to determine blowdown emissions.
98.233 (i)(3): CAPP recommends that equation W-10 be a summation of all individual
blowdown events as the volume blown down will change because it is unlikely that the whole
piping assembly will be blown every event. Additionally the summation would be required to
determine the "average" volume.
Response: EPA is not clear on the commenters request for change. If the commenter has
multiple sections of the system that are blow down at different intervals then EPA has made the
necessary changes to the equation to allow for summation of different blowdown volumes. If
unique chamber volumes are blowndown in separate events, then the volume between isolation
valves will have to be calculated separately for each unique volume. In addition, EPA requires
reporting of total annual venting emissions for each equipment type as specified under Section
98.233(i)(3) of todays final rule.

1505
Errors in Section Labeling Should Be Corrected

The Proposed Rule includes a typographical error in the section hierarchy for the blowdown
calculation section. The Proposed Rule indicates this subsection is 98.233(h)(iii); this
subsection should be 98.233(i). This error is not found in the pre-publication version of the
Proposed Rule, but is in the Federal Register version of the Proposed Rule. There is a related
section hierarchy error in 98.233(h), where the first subsection is labeled (h)(i). The proper
subsection protocol is to migrate from letters to numbers to (i), (ii), etc. Thus, 98.233(h)(i)
and (ii) should be labeled 98.233(h)(1) and (2).
For clarity, INGAA provides appropriate revisions to the erroneous 98.233(h)(i) (iii) text
from 75 FR 18640:
(i)(1) Calculate natural gas volumetric emissions at standard conditions using calculations in
(iii)(2) Both CH4 and CO2 volumetric and mass emissions shall be calculated from volumetric
(iii)(i) Blowdown vent stacks. Calculate blowdown vent stack emissions as follows:
Response: EPA agrees with the comment about the typographical error in the middle column
of 75 Fed. Reg. 18640 and has made necessary corrections to todays final rule.

Emission Calculation Flexibility Should Be Provided
Subpart W should allow flexibility in documenting the emission calculations. Many companies
currently employ algorithms and programs to calculate and document blowdowns, and these
methodologies use the same ideal gas law principles as intended by the Subpart W equations to
calculate event-based blowdown volumes. Companies currently employing analogous
equivalent methods would document the calculation basis and Subpart W equivalence in their
GHG Monitoring Plan. This could include an approach that calculates the gas volumes for
individual venting events and sums the event volumes to determine annual emissions. For
blowdowns of equipment where process conditions (e.g., blowdown beginning and/or ending
pressures) vary for different events that occur in a year, this approach would provide a more
accurate emission estimate than the Proposed Subpart W equations which assume the process
pressure change and temperature are the same for each blowdown
Response: After considering several comments, EPA has made modifications to the blowdown
methodology to allow for the estimation of blowdown volumes with differing temperature and
1506
pressure conditions. With regard to these modifications in equations for blowdown vent stacks,
The commenter has not provided sufficient details on the other methods that it intends to use.
Hence EPA does not have any comment on these other methods. Finally, EPA determined that
the engineering calculation methods in todays final rule are sufficiently simple and low cost to
estimate and report blowdown emissions. Hence, EPA has not allowed for any new methods to
estimate blowdown emissions.
Equations for Blowdown Volume and Volume Correction to Standard Conditions Require
Revisions
Corrections to equations and nomenclature presented in 98.233 are needed to correct
calculations and to properly represent blowdown pressure changes and associated volumes:
As currently written in 98.233, Ea,n in equation W-10 should be defined as, Annual natural
gas venting emissions at ambient process conditions from blowdowns in cubic feet. That is, the
gas inside the vented equipment is at process temperature and pressure. The gas is not at ambient
conditions.
The 98.233 pressure correction calculation in Equation W-20 is not correct for blowdown
emission estimates. The pressure parameters are ambient and standard conditions in
Equation W-20, whereas the relevant pressures are ?P the system/process gas pressure change
during the blowdown and standard pressure. That is, the system pressure change from event
start to finish is not the same as ambient and standard in Equation W-20, and the equation to
correct volumes to standard conditions requires revision.
Thus, Equation W-10 could be revised to eliminate a separate calculation (i.e., Equation W-20)
to correct to standard conditions via the following formula: [ See original comment for equation]
Where:
Es,i = volume of gas vented, at standard temperature and pressure, from event i;
Vi = total geometric volume of blowdown equipment chambers (including, but not limited to,
pipelines, compressors and vessels) between isolation valves in cubic feet;
PI = initial equipment gas pressure; i.e. gas pressure at the start of the blowdown;
PF = final equipment gas pressure; i.e. gas pressure at the end of the blowdown;
PSTD = standard gas pressure (14.7 psi);
TSTD = standard gas temperature (60F);
TP = process gas temperature (F); and
Es = annual volume of gas vented, at standard temperature and pressure, from the equipment.
1507
With the above revisions, the result is at standard conditions and Equation W-20 would not be
needed to calculate emissions at standard conditions. Note that these estimation equations do not
consider the non-ideal gas behavior of natural gas at various pressures (i.e., equations do not
include the compressibility factor z) and assume the process temperature does not change
during the venting (i.e., the initial and final process temperatures are the same). These are
standard assumptions for calculating vented volumes. In general, the effects from these
parameters will typically be small relative to pressure changes, but in some cases may be
included in company algorithms and programs for more rigorous calculations.
If EPA makes alternative revisions to Equation W-10 to determine vented volumes at other than
standard conditions and retains the reference to Equation W-20, then Equation W-20 must be
revised so that corrections to standard conditions can consider the scenario pressure and
temperature, which may be at process conditions or ambient pressure and temperature. Currently,
Equation W-20 only provides a means to correct from ambient conditions to standard conditions.
While INGAA has not reviewed requirements for all industry segments in the Proposed Rule,
INGAA expects there are additional rule references to 98.233(t) and Equation W-20 that reflect
similar anomalies. EPA should reconcile other anomalies or reference conditions (e.g., process
versus ambient conditions for volume calculations) that introduce errors
Response: EPA agrees with the commenter on the errors in the equation to estimate blowdown
emissions. In todays final rule, EPA has corrected the equation to use inputs at actual conditions
and provide an output at STP conditions. EPA would like to note that the compressibility factor
is 1 because of the assumption that ideal gas law holds at actual operating conditions. This
approximation will not result in any significant errors in calculations, but will simplify the
calculation method. Please see response to EPA-HQ-OAR-2009-0923-1011-41 for further
clarifications in the monitoring method.

Safety Must Be Considered and Alternatives Provided for Inaccessible Vents
Safety is an extremely important concern when attempting to collect measurements from roofline
vents at transmission compressor stations. At existing facilities, most vents are routed outside the
compressor building and elevated above the roof line to disperse vented gas. Similarly, there are
safety concerns associated with vent access for condensate tank vents at some facilities.
Resolving these safety concerns will likely require modifications to provide vent access at a safer
location. However, line accessibility (e.g., within the compressor building) may not be readily
available as the facilities were not constructed considering these criteria. Due to an aggressive
proposed implementation schedule, this would put operators at risk of failing to comply with the
annual survey requirement, or placing personnel in harms way i.e., violating safety procedures.
Additional time should be provided to implement the vent measurement program so that
modifications can be made to accommodate measurement from a safer location than from a
1508
building roofline or difficult to access condensate tank vent.

Safety concerns and vent measurement issues are exhibited in a photograph from a recent pilot
study shown in Figure 1. To reach the roofline vents, the picture shows the manlift extended to
maximum length and precariously suspended above high-pressure natural gas lines. This
measurement should not have been pursued and cannot be mandated, as it creates unsafe
conditions for workers and our facilities. In addition, vent locations require manlift placement in
areas not designed for that use, resulting in significant destruction of walkways and the pipeyard
grounds at this facility. For this pilot study, even with unacceptable manlift use, not all vents
could be accessed and were therefore not measured.
Figure 1. Manlift Use for Roofline Vent Access Can Introduce Significant Safety Concerns.[See
original comment for Figure 1]
In addition to the safety issues clearly shown in the above figure, vent measurement also
introduces safety concerns by placing personnel in proximity to vents that are potential release
points if an unplanned emergency blowdown were to occur. Even if safe access is achieved,
additional safety concerns exist regarding roofline vent measurement. Facility safety
requirements preclude the testing of a functioning pressurized compressor relief valve vent that
could potentially discharge during measurement activities, thus placing personnel in a potentially
flammable or explosive environment. To address this issue, process and procedural
modifications would be required to eliminate the potential for an emergency blowdown.
Implementation would be costly and complex.
The majority of building roofline vents and some condensate tank vents would be deemed
inaccessible under conventional LDAR programs, as such components would be considered
unsafe or difficult-to-monitor in a typical LDAR program (e.g., components that cannot be
monitored without elevating the monitoring personnel more than 2 meters above a support
surface). Under a typical LDAR program, monitoring can be deferred due to unsafe or difficultto-monitor components. In response, the facility must maintain documentation that explains the
conditions under which the components become safe to monitor or no longer difficult to monitor.
For vent lines at most compressor stations, typical conditions under which these vents would be
safe would preclude normal engine operations or pressurized scenarios where the potential for
a vented release is possible. References are available from several rules that require VOC
reductions through LDAR programs. Examples include the following:
40 CFR 60, Subpart VVa addresses performance standards for VOC leaks from chemical
manufacturing and provides criteria for addressing access and safety issues including these
examples:
- 60.482-7a(g) identifies a valve as unsafe to monitor if:
The owner or operator of the valve demonstrates that the valve is unsafe to monitor because
monitoring personnel would be exposed to an immediate danger as a consequence of
complying [with leak monitoring requirements];
1509
- 60.482a-11a(f) identifies inaccessible and unsafe connectors:

(f) Inaccessible, ceramic, or ceramic-lined connectors . (1) Any connector that is inaccessible
is exempt from the monitoring requirements and from the recordkeeping and reporting
requirements . An inaccessible connector is one that meets any of the provisions specified in
paragraphs (f)(1)(i) through (vi) of this section, as applicable:
(iii) Obstructed by equipment or piping that prevents access to the connector by a monitor probe;
(iv) Unable to be reached from a wheeled scissor-lift or hydraulic-type scaffold that would allow
access to connectors up to 7.6 meters (25 feet) above the ground;
(
v) Inaccessible because it would require elevating the monitoring personnel more than 2 meters
(7 feet) above a permanent support surface or would require the erection of scaffold; or
(vi) Not able to be accessed at any time in a safe manner to perform monitoring. Unsafe access
includes, but is not limited to, the use of a wheeled scissor-lift on unstable or uneven terrain, the
use of a motorized man-lift basket in areas where an ignition potential exists, or access would
require near proximity to hazards such as electrical lines, or would risk damage to equipment.
60.486a(f) includes provisions for documenting and retaining records for components that are
unsafe or difficult to monitor.
There are additional examples from similar performance standards for VOC leaks. These criteria
regarding inaccessible and unsafe-to-monitor components apply to many vents at transmission
and storage facilities, and it is clear that regulated VOC sources have allowances for inaccessible
components. Similar criteria should be applied to address vent access under Subpart W.
Implementation also requires additional training and expertise. Access to elevated vents requires
test personnel to obtain fall protection training and annual certification. Physical access to the
majority of reciprocating compressor rod-packing vents requires a manlift or a ladder on an
elevated walkway grate that is typically located above high pressure discharge yard piping. A
certified manlift operator is required to safely position the manlift bucket to properly access and
measure vented gas volumes. Facility grounds in proximity to these locations may not be
conducive to manlift siting, and care must be taken to avoid uneven ground, drop offs, holes,
adjacent structures, pipe racks, and overhead obstructions. Weather and surface conditions can
also significantly affect the ability to properly position the manlift. These access restrictions
present a considerable safety risk, which could lead to reassessment of insurance liability. In
addition, facility and tank configuration for gas transmission condensate tank vents may present
similar access problems at some facilities. Clearly, there will be instances when vent access is
not possible and/or would violate accepted safety practices.
INGAA members are currently pursuing alternative sample points at safer vent measurement
locations. However, accessibility will be an issue at some facilities, and this will not be a trivial
matter when Subpart W is implemented across the nations transmission pipeline system. There
1510
is no reasonable basis to impose an implementation schedule that requires operating a manlift

unsafely just to acquire a mandated measurement.
As discussed in Comment XII, additional time should be allowed to implement Subpart W.
Furthermore, allowances must be provided in the Final Rule if facility design precludes
reasonable access to vent lines or condensate tank vents at a safe measurement location.
Operators should not be required to pursue extraordinary facility modifications, such as costly re
design and modification to existing infrastructure, only for the purpose of vent access. In these
instances, operators should be able to identify inaccessible vents within the GHG Monitoring
Plan, as well as include the basis for such judgments and explanation as to the infeasibility of
alternative vent line access. Best available data options should be allowed for emissions
estimates under these cases. INGAA offers its assistance to work with EPA to define a
reasonable basis to ensure inventory integrity, while considering important safety issues
regarding vent accessibility
Response: EPA considered all methods for quantifying equipment leaks and vents. All methods
proposed in todays final rule have been demonstrated in practice in operating facilities, and are
considered safe to perform by trained technicians. EPA agrees with the commenter that safety
must be a first consideration in all emissions measurement activities and in certain unique cases,
the methodologies as outlined in the proposed rule may pose a safety hazard. Therefore, for
compressor vent emissions where roof vents are not positioned in a location that is safe to access
the end of the pipe, the rule allows for a piping modification to install a port for insertion of a
flow measuring instrument, or installation of a permanent flow measurement instrument. In those
cases where this modification cannot be made for a particular equipment configuration, the
reporter may petition EPA for an extension of time under the BAMM provisions in todays final
rule. Please see Section II.F of the preamble to todays final rule for further details.
EPA also agrees with the commenters point about inaccessible emissions sources from
condensate tanks in gas processing facilities and transmission compressor stations. In todays
final rule, EPA has determined that condensate tanks in processing facilities do not contribute
enough emissions to warrant the cost of quantification, and therefore this source is not required
to be reported.
In gas transmission compressor stations, EPA has included in todays final rule the alternative of
using an acoustic leak detection instrument that has algorithms for through-valve leak
quantification to estimate compressor scrubber dump valve leakage as an alternative to
measuring the emissions from the tank roof vent. This alternative acoustic through-valve leak
measurement technique is also added in todays final rule for compressor blowdown vent
through-valve leakage. Because the blowdown vent valve must be accessible to operate, this
acoustic detection alternative would be more cost-effective and safe to employ for this emission
source quantification.

1511

Blowdown Venting
Subpart W annual emission estimates for equipment blowdowns are calculated from the volume
of gas vented during each event and the number of events per year. The volume of gas vented is
determined from the vented equipment internal geometric volume with corrections for pressure
and temperature to standard conditions.
In general, INGAA accepts the proposed method in 98.233(i) for estimating equipment
blowdown emissions, but INGAA recommends minor revisions to add clarity and correct
erroneous references and formula errors the Subpart W calculation of vented gas volume at
standard conditions (equation W-10 and equation W-20) is not correct and requires revision. In
addition, flexibility in documenting the emission calculations is needed.
Response: EPA has made modifications to the blowdown methodology to allow for the
estimation of blowdown volumes with differing temperature and pressure conditions. With
regard to these modifications in equations for blowdown vent stacks, please see the response to
EPA-HQ-OAR-2009-0923-1011-41.

Blowdown vent stack requirements should be clarified. In the middle column
of 75 Fed. Reg. 18640 there are requirements set forth for blowdown vent stacks. Those
provisions are numbered (iii) but in fact should be numbered (i). There is currently a lack of
clarity as to what items must be subjected to calculation pursuant to these provisions, given the
proposal's ill-advised use of "including but not limited to" in the applicability section of the rule.
There are hundreds of items that could be covered by a rule that covered calculation of "total
volume . . . between isolation valves," such as sample bottles, sight glass, control tubing, and the
like. We expect that EPA does not intend for such de minimis sources to be subject to
calculation, and we suggest that the rule be amended to provide for a de minimis level of 5 mcf
below which calculation of individual items such as mentioned above need not be performed. In
addition, we believe that subparagraph (1) should be amended as follows:
"(1) Calculate the total volume (including, [strike through: but not] limited to, [strike through:
pipeline] process piping, compressor case or cylinders, manifolds, suction and discharge bottles
and vessels) between isolation valves."
These changes will add clarity for regulated companies and will eliminate unnecessary and
onerous reporting of very insignificant sources.
Response: EPA agrees with the comment. Todays final rule provides an equipment threshold
below which emissions do not have to be reported. Please see response to EMAIL-0002-8
1512
OAR-2009-0923) for further details and the memo on the Equipment Threshold for Small
Combustion Units under docket EPA-HQ-OAR-2009-0923.
In todays final rule, EPA has also accepted the correction suggested by the commenter to the
calculation of total volume between isolation valves.
EPA also agrees with the comment about the typographical error in the middle column
of 75 Fed. Reg. 18640 and has made necessary corrections to todays final rule.

Commenter:
Comment: WBIH strongly requests measurement of emissions from rod packing, blowdown
valves, and blowdown vents be conducted in one mode, the mode existing at the time of the
measurement.
In order to conduct measurements in the compressor de-pressurized mode, a compressor

blowdown would be required, resulting in additional emissions. Additionally, in the compressor
de-pressurized mode, the compressor is not available for service creating an interruption of
service to customers
Response: EPA never intended for operators to depressurize their compressors to estimate
emissions in that mode. Please see the response to EPA-HQ-OAR-2009-0923-0055-16 for
further details on EPAs revised monitoring requirements for compressors.

Commenter:
Comment on 98.233 (75 FR 18640) for "Blowdown vent stacks":
Correct the identification reference from "iii" to "i".
Response: EPA agrees with the comment about the typographical error in the middle column of
75 Fed. Reg. 18640 and has made necessary corrections to todays final rule.

1513
Aka recommends that GHGs from only the following blowdowns be reported under 98.233(i):
(1) all compressor blowdowns; and (2) blowdowns from process vessels and piping with a
physical volume of 50 cubic feet and larger.
Response: EPA agrees with the commenter on reporting all compressor blowdowns and on
providing an equipment threshold. With regard to blowdown vent stack emissions requirements
and equipment threshold, please see the response to EMAIL-0002-8 (comment also located in

Section 98.233(i) Blowdown vent stacks. In Section 98.233(i)(2), logs of the number of
blowdowns for each equipment type are required to be maintained. While a record of blowdown
occurrences is usually kept for larger types of equipment such as compressors greater than 500
hp, keeping records of blowdowns for the numerous small pieces of equipment is impractical.
API recommends engineering estimates be used for the number of blowdowns for compressors
less than 500 hp and equipment other than compressors. The engineering estimate methods
would be described in the monitoring plan.
Response: In todays final rule EPA has modified blowdown vent stacks emissions requirements
and provided an equipment threshold. For further clarification, please see the response to
EMAIL-0002-8 (comment also located in rulemaking memo Early Comment Submissions in
docket EPA-HQ-OAR-2009-0923).

Blowdown Vents (Natural Gas or CO2 Enhanced Oil Recovery Operations)
We support EPAs proposed emission estimating method for blowdown vents that is based on
the number of blowdowns and the volume of the blowdown equipment chamber.
Response: EPA has made a minor modification to the blowdown methodology to allow for the
estimation of blowdown volumes with differing temperature and pressure conditions and also
provided an equipment threshold. Please see the responses to EPA-HQ-OAR-2009-0923-1011
41 and EMAIL-0002-8 (comment also located in rulemaking memo Early Comment
Submissions in docket EPA-HQ-OAR-2009-0923) for further details.
1514

98.233(e)(3): This requirement addresses emissions during depressurization of a solid media
dehydration vessel. These emissions are already covered by Section 98.233(i), and this section
should be deleted to avoid double reporting. Also, there is a typo on page 18640. The section that
should be labeled 98.233(i) Blowdown Vent Stacks is mislabeled as 98.233(h)(iii).
Response: EPA assumes that by solid media dehydration vessel the commenter means
dessicant dehydrators, in which case EPA agrees that the April 2010 proposed rule found in
docket (EPA-HQ-OAR-2009-0923-0002) would have resulted in double counting of emissions.
EPA has clarified in todays final rule that desiccant dehydrators must report under the
dehydrator vent source type.
EPA agrees with the comment about the typographical error in the middle column of 75 Fed.
Reg. 18640 and has made necessary corrections to todays final rule.

Blowdown vent stacks
Section 98.233(i): There are a limited number of tasks in the natural gas industry that result in
blowing down significant volumes of natural gas to the atmosphere. These activities are limited
to blowing down compressors, process vessels, and piping. There are other tasks in the natural
gas industry that result in blowing down very small and insignificant volumes of natural gas,
such as blowdowns from sight glasses and small tubing. Without thresholds that limit the
blowdowns that must be recorded, calculated, and reported, this portion of Subpart W is
unmanageable. IPAMS requests that only GHG emissions from the following types of
blowdowns be reported: 1) compressor blowdowns, and 2) blowdowns from process vessels and
piping with a physical volume of at least 50 cubic feet.
Also note that this section was incorrectly positioned in the proposed rule as Section
98.233(h)(iii); the correct section should be labeled as Section 98.233(i).
Response EPA does not agree with the commenter on limiting blowdowns to compressors only.
However, EPA agrees with the commenter on providing an equipment threshold. Please see
response to EMAIL-0002-8 (comment also located in rulemaking memo Early Comment
Submissions in docket EPA-HQ-OAR-2009-0923) for further details.
1515
EPA agrees with the comment about the typographical error in the middle column of 75 Fed.
Reg. 18640 and has made necessary corrections to todays final rule.

Section 98.233(i) Blowdown vent stacks
In Section 98.233(i)(2), logs of the number of blowdowns for each equipment type is required to
be maintained. While a record of blowdown occurrences is usually kept for larger types of
equipment such as compressors, keeping records of blowdowns for the numerous small pieces of
equipment is impractical. BP recommends engineering estimates be used for the number of
blowdowns for equipment other than compressors.
Response: After considering several comments, EPA has modified blowdown vent stacks
monitoring requirements and provided an equipment threshold that will mitigate commenter
concern for non-large equipment blowdown emissions. However, EPA does not agree that the
source be limited to compressors only. For further details, please see response to EMAIL-0002-8
OAR-2009-0923).

Blow-down Systems: A Blow-down system is typically designed at an elevated height in order to
safely vent high pressurized gas away from the gas facility which can be done when compressors
are placed offline due to changing operational demands or for purposes of conducting
maintenance. The areas of venting where the blow-down event occurs can be 12 60 feet off the
ground and not easily accessible. Gas may also be escaping through these stacks or vents because
of leaking seals whereby internal valve leakage can also vent safely to atmosphere. Conventional
methods used to screen these vent systems have included using a long extension probe
(telescopic, see figure 1 below) to pull a sample of air into a pump driven combustible gas
indicator (using catalytic oxidation / thermal conductivity hydrocarbon sensors) or infrared leak
detectors in order to determine gas concentration as a percentage of volume gas or parts per
million. The gas imaging camera does offer some advantage in screening these difficult to reach
systems because you can view the stack from the ground and make a determination from the
image viewed in the display. However, the gas imaging camera does not offer any additional
information on the gas leak such as gas concentration or leak rate.
The gas imaging cameras costs anywhere from $85,000 $105,000 and are heavily dependent on
1516
user experience and expertise. There is a looming dilemma with the use of these cameras in that
the utilization of them becomes more subjective from user to user. What one user may see
through the lens of a camera, another may completely miss because they did not adjust the image
to the proper polarity and enhance the sensitivity. Weather conditions, temperature and
cloudiness will affect whether or not a gas plume can be properly and adequately imaged which
invite potential problems for gas companies who may later be audited and asked to account for
missed gas leaks.
Other pitfalls with the imaging camera include the question on what to do with the stored image
files and how to properly manage this piece of information. It will take a great amount of time
and attention to detail in tracking this information and applying it to a database against a known
or clearly identified component in the system.
Operators will be subject to fatigue with this device as it requires enormous mental focus and
physical strength to first stabilize the camera, scan the area, adjust the settings and determine
whether a leak plume is present or not.
Most of these considerations are removed with the use of readily available, proven, much more
affordable leak detection technologies which are manufactured not by one single company, but
multiple manufacturers in the United States and abroad. A picture below is provided as an
example of a safe, reliable, accurate and fast method for screening blow-down systems which
may include vents/stacks designed for blow down valves, main-line suction/discharge block
valves or unit valves, pressure relief valves, power gas starter valves and emergency blowdown/shutdown valves.
Figure 1: Vent stack leak screening with an electronic gas leak detector, TVA/OVA or
infrared gas detector. [See pdf attachment for photo]
Response:
EPA does not require leak detection for the blowdown emissions source type (Section 98.233(i)
of todays final rule). Hence, the comment is irrelevant in the context of this source type. EPA
requires the use of engineering estimation of blowdown emissions.
1517
13.2.7
COMPRESSORS

Monitoring methods and emissions quantification for: Centrifugal Compressor Wet Seal
Degassing Vents and Reciprocating Compressor Rod Packing Venting. The effort required to
measure the emissions from centrifugal compressor wet seals and reciprocating compressor rod
packings is extensive. The vents from these two pieces of equipment are combined with the vents
of other pieces of equipment and are emitted from the blowdown vent stack away from the
compressor. The individual vents for these two pieces of equipment are very difficult to access
inside the compressor and the emissions from the blowdown vent are combined with other
emissions. Breaking these vent headers apart inside the compressor building for measuring
purposes can pose a safety hazard. It would be very difficult to measure emissions from the
blowdown vent stack because it ranges from 1,000's of cubic factions of feet of gas per hour. The
volume going thru the venting system when starting and stopping the compressor is very
different when compared to times when the compressor is at idle. It would be very difficult to
identify the emissions from a single particular piece of equipment. In lieu of measuring the gas
from centrifugal compressor wet seals and reciprocating compressor rod packings, NMGC
suggests requiring that companies demonstrate good maintenance practices of these pieces of
equipment. If properly maintained, the emissions are very small. Requiring documentation of
proper maintenance, i.e. regularly replacing them, is an incentive for companies to maintain the
equipment properly which in turn will ensure emissions are minimal.
Response: EPA does not require compressors to be shutdown. Please see Section II.F of the
preamble to todays final rule and Compressor Modes and Threshold (EPA-HQ-OAR-2009
0023). EPA does not require that vent manifolds be broken apart. Please see response to

Section 98.233 (p) (4) Conduct one measurement for each compressor in each of the operational
modes that occurs during a reporting period: Our experience has shown that most of the large
leaks found on compressors have occurred during operating mode. To image and measure all
vents in all three operational modes will create a number of challenges associated with bringing
compressors down and then back up which will result in excessive releases of methane, as well
as the required man-hours associated with having the optical imaging crew on standby while
compressors are brought down and back on-line.
1518
Response: EPA disagrees with this comment. For more information please see Section II.F of
the preamble to todays final rule and Compressor Modes and Threshold (EPA-HQ-OAR
2009-0023).

Organization:Yates Petroleum Corporation
Commenter:
General Comments Fugitive Emissions at Different Operational Modes. On page 43 of the
Technical Support Document and in 98.233(p)(4), the EPA addresses fugitive emissions from
compressor operations at different operational modes. In the Technical Support Document, the
EPA concedes that to address this issue, operators must measure emissions for each mode the
compressor is operated in and the duration of that operational mode and this will increase the
reporting burden, since measurements will have to be taken at each mode of compressor
operation. Indeed this requirement will greatly increase the reporting burden while not
addressing a significant source of GHG emissions. 98.233(p)(4) requires measurement in several
operational modes: Compressor engines are generally either operational or not. Some of the
operational modes are not normally encountered and this would have to be artificially
reproduced in the field for the sole purpose of reporting. A compressor would not standby
pressurized for any amount of time that would affect its emissions significantly as it would
otherwise be offset by the fact that the compressor is operational. Therefore, requiring
measurement in each mode is burdensome, difficult to schedule, and does not reflect a significant
source of emissions.
2009-0023).

Aka recommends that industry not be required to measure emissions from standby or
depressurized compressor units. These measurements will require the creation of GHG emissions
in the depressurization of units and will add no useful data to the data set for most units.
2009-0023).
1519

Compressor Emissions. We support EPAs proposed emission estimating method for
reciprocating and centrifugal compressors that relies on direct measurement.
Response: EPA reviewed this this comment and has retained direct measurement of compressors
in the rule.

In 98.233(r)(2) CAPP questions the inclusion of "compressor starter gas vents" in equation W
19. Is the intent to capture methane emissions during starting events? If so, these emissions are a
function of the number of starts per year (duration of only a few seconds each). If the intent is to
capture emissions from periods when the engine is running and gas is passing through the starter
to atmosphere, the calculation should reflect the calculation for an open-ended line as this is what
most accurately represents the emission event.
Response: Upon further review of the large emissions sources as described in the Technical
Support Document EPA-HQ-OAR-2009-0923, EPA has determined that compressor starter gas
vents are not a significant source of GHG emissions. Therefore, EPA has removed compressor
starter gas vents from the list of sources.

Section 98.233(r)(2) Population count and emission factors. API would like clarification on the
inclusion of compressor starter gas vents under Equation W-19. Emissions from these vents
are not a function of compressor operational hours, but rather are dependent on the number of
starts per year whose duration only last a few seconds. In addition, this section refers to Table
W-1 for a compressor starter gas vent emission factor, but such a factor is not listed in Table W
1. This emission source is also referred to as a fugitive source, but it is considered a vented
emission source in the API Compendium.
Response: EPA has removed compressor starter gas vents from todays final rule. Please see
1520

Reporting compressor throughput: This section requires companies to report throughput for each
compressor covered by Subpart W. Compliance with this requirement would require a flow
meter to be installed on every affected compressor in the E&P sector. This requirement would be
excessively costly because individual compressors typically are not equipped with flow meters,
so countless new meters would have to be installed. Moreover, individual compressor throughput
is data that is not useful to EPA for any reasonable policy or regulatory purpose. Thus, we
recommend that this reporting requirement should be deleted. Alternatively, if EPA determines
that compressor throughput is needed, that information should be provided on a per facility basis
(i.e., as an aggregate for all compressor at a given facility rather than for each compressor
individually) and throughput should be based on engineering estimates rather than direct
measurement.
Response: Commenters made several assumptions which were not consistent with the proposed
rules intent. One was the assumption that each compressor would need a meter to determine
compressor throughput. Reporters are not required to install a meter. EPA has clarified this in
todays final rule and allows reporters to use engineering estimation to determine compressor
throughput. EPA requires the compressor throughput for analysis of the activity data and the
resultant GHG emissions reports, as combustion CO2 will be proportional to compressor
throughput.

Commenter: Kim Dang
Reporting of Individual Compressor Throughput. The proposed rule appears to require reporting
of total throughput of each individual compressor connected to a compressor wet seal degassing
vent, as well as the total throughput of individual reciprocating compressors. 220 Kinder Morgan
and other members of our industry generally do not monitor throughput at the level of individual
compressors; rather, this data is collected at the compressor station level. Obtaining this data
would be costly and difficult. Furthermore, Kinder Morgan believes this data would have little
practical use for EPA, since the quantity of vented and fugitive emissions from compressors is
not well correlated with throughput.
Response: EPA agrees that many companies do not monitor throughput and therefore EPA has
modified to todays final rule to allow the use of engineering estimates. EPA disagrees with the
220
Proposed 40 C.F.R. SECTION 98.236(c)(17)(iii), (18)(i).

1521
comment that the data would have little practical use for EPA. For further information on both
points, please see response to comment EPA-HQ-OAR-2009-0923-1206-63.

LNG Compression Equipment. LNG storage facilities use reciprocating and centrifugal
compressors for a limited number of purposes, which include boil-off compression, and as
refrigerant and boost compression in liquefaction processes. Compressors are not used to
compress vaporized LNG to move to the send-out system, as described in the EPAs Background
Technical Support document (see Exhibit C, Figures 1A, 2, and 3).
While reciprocating compressors at LNG storage facilities may appear similar externally to
reciprocating compressors more familiar to EPA, such as those located at transmission
compressor stations, compressors at LNG storage facilities are typically quite different
internally, due to the materials utilized to seal piston rod packing to limit leakage past the piston
rods. Reciprocating compressors used at LNG storage facilities for boil-off compression,
liquefaction refrigeration compression and liquefaction stream boost generally utilize nonlubricated cylinders and rod packings to eliminate the risk of lubricants being introduced into the
process gas stream causing contamination of the stream with oil residue, which would freeze the
oil in critical components such as heat exchangers, operating at temperature as low as -260
degrees Fahrenheit. These non-lubricated applications typically utilize piston ring, and rod
packing materials of varying graphite, PTFE (Teflon) blends which are very resilient and have
high sealing qualities, thereby greatly minimizing fugitive leak emissions from the compressor
rod packing cases.
Centrifugal compressors are typically utilized at LNG storage facilities as liquefaction process
refrigerant compressors. Again, while externally they may resemble typical centrifugal
compressors that are becoming more common in gas transmission compressor stations, these
centrifugal refrigerant compressors are typically quite different internally with respect to their
shaft sealing design, using very sophisticated oil film (wet) sealing or dry elements, which
greatly reduce or eliminate leakage. The outer case seal areas and the seal oil drains are directed
back to the compressor suction after demisting takes place. Because the outer case seal area and
seal drains lead back to the compressor suction, there are no gases purposely vented to the
atmosphere as the proposed rule assumes. All of the above discussion on compressors provides
further evidence that LNG facilities are not contributing to the fugitive emissions concerns that
this proposed rule is intended to minimize.
For example, reciprocating compressor rod packing has a certain level of emissions by design,
e.g., there is a clearance provided between the packing and the compressor rod for free
movement of the rod that results in emissions. Also, by design, vent stacks in petroleum and
natural gas production, natural gas processing, and petroleum refining facilities release natural
gas to the atmosphere. Unintentional emissions result from wear and tear or damage to the
equipment. For example, valves result in natural gas emissions due to wear and tear from
1522
continuous use over a period of time. Also, pipelines damaged during maintenance operations or
corrosion result in unintentional emissions.
Response: EPA disagrees with this comment. EPAs decision to include LNG storage facilities
was based on several considerations. Please see EPAs Greenhouse Gas Emissions from the
Petroleum and Natural Gas Industry: Background TSD (EPA-HQ-OAR-2009-0923-0027) and
response to comment EPA-HQ-OAR-2009-0923-1025-1 for EPAs analysis for including LNG
storage in the rule. In addition, EPA is aware of different technologies for compressors deployed
by industry across several sectors to reduce emissions. If this commenter is using low emission
technologies, then EPA would expect to see such emissions reflected in their submitted report.
However, the use of low emission technology is not known to be ubiquitous in the LNG industry
and therefore the collection of GHG data from the LNG sector is important in understanding
emissions from this sector and therefore does not entitle LNG storage facilities special
considerations under this economy-wide GHG rule; as such information may demonstrate
differentiated emissions levels, which would inform future policy.

Gathering compression facilities are typically small facilities and widely dispersed. Many of
these compression facilities have only one or two compressors and may include treatment
facilities, such as dehydration units. Significantly-sized facilities are already subject to Subpart C
reporting if the combustion emissions exceed 25k tonnes per year (tpy) and individual
compression facilities that exceed 25k tpy, including Subpart W emissions, would also begin
reporting as required. The proposed Subpart W, however, treats these small and widely dispersed
facilities as if they are the same size and complexity of a gas processing plant. Examples of
proposed requirements that may be appropriate for processing plants but unduly burdensome for
small and widely dispersed gathering compression facilities include: collecting extensive data
and modeling tank emissions, compressor rod packing vents, leak detection using optical
imaging or population factors, and quarterly sampling of gas streams.
Response: Todays final rule does not include gathering and boosting stations. Please see
Section II.F of the preamble to todays final rule.
13.2.7.1
CENTRIFUGAL COMPRESSORS

Section 98.233 (o) Centrifugal compressor wet seal degassing vents. This section states that a
1523
flow meter, either temporary or permanent, is to be used to measure emissions from ALL
VENTS to the atmosphere or to the flare. If the compressor vent is connected to the flare, and
the flare has a gas meter attached, this would result in double counting. We would also suggest
that the optical imaging camera be used to first determine if the vent is leaking before moving
forward. This would potentially save the expense of measurement, and would eliminate the
safety hazard of using a man lift in and around the components.
Response: Commenters made several assumptions which were not consistent with the proposed
rules intent. One was the assumption that EPA required the double reporting of flare emissions
calculated under each paragraph of the rule in 98.233 Calculating GHG emissions, that
requires flare emissions determination. Reporters are not required to double report flare
emissions. EPA has clarified this in todays final rule and states that reporters must correct flare
emissions to avoid double reporting. In addition, centrifugal wet seal degassing tank venting is
by design, and therefore always occurs when the high-pressure oil barriers for centrifugal
compressors are depressurized to release absorbed natural gas. High-pressure oil is used as a
barrier against escaping gas in centrifugal compressor shafts. Very little gas escapes through the
oil barrier, but under high pressure, considerably more gas is absorbed by the oil. The seal oil is
purged of the absorbed gas (using heaters, flash tanks, and degassing techniques) and
recirculated. Therefore, EPA disagrees with requiring the use of an optical gas imaging camera
to detect seal oil degassing tank venting, because when the separated gas is vented to the
atmosphere, by design, gas will be venting.

It appears that you must calculate the volume of vapors sent to a flare or tank for centrifugal
compressor wet seal degassing vents. Would this not be also reported under flares? Is this
double reporting of emissions
Response: EPA disagrees with this comment regarding flare double reporting. Please see
response to comment EPA-HQ-OAR-2009-0923-0133-4 above.

USEPA Subpart W approach: The re-proposed Subpart W reporting regulation contains methods
for both reciprocating and centrifugal compressors. These two methods are discussed below.
Centrifugal compressor wet seal degassing vents
Calculate emissions from centrifugal wet seal degassing vents as follows:
Step 1: For each centrifugal compressor determine the volume of vapors from the wet seal oil
1524
degassing tank sent to an atmospheric vent or flare, using a temporary or permanent flow
measurement meter, such as, but not limited to, a vane anemometer, according to the methods set
forth in 98.234(b).
Step 2: Estimate annual emissions using meter flow measurement using Equation W-16 of this
section.
E [subscript a, i] = MT * T * M [subscript i] * (1 ) (Eq. W-16)
Where:
E [subscript a, i] = Annual GHG I (either CH4 or CO2) volumetric emissions at ambient
conditions.
MT = meter reading of gas emissions per unit time
T = total time compressor associated with the wet seal(s) is operational in the reporting year.
M [subscript i] = mole percent of GHG i in the degassing vent gas; use the appropriate gas
composition in paragraph (u)(2) of this section
= percentage of centrifugal compressor wet seal degassing vent gas sent to vapor recovery or
fuel gas or other beneficial use as determined by keeping logs of the number of operating hours
for the vapor recovery system and the amount of vent gas that is directed to the fuel system
Step 3: Calculate CH4 and CO2 volumetric emissions at standard conditions using paragraph (t)
of this section.
Step 4: Calculate both CH4 and CO2 mass emissions from volumetric emissions using
calculations in paragraph (v) of this section.
Step 5: Calculate emissions from degassing vent vapors to flares as follows:
(i) Use the degassing vent vapor volume and gas composition as determined in paragraphs (o)(1)
through (3) of this section.
(ii) Use the methodology of flare stacks in paragraph (n) of this section to determine degassing
vent vapor emissions from the flare.
Reciprocating compressor rod packing venting. Calculate annual CH4 and CO2 emissions from
each reciprocating compressor rod packing venting as follows:
Step 1: Calculate CH4 and CO2 volumetric emissions at ambient conditions as follows:
E a,i = MTT M (Eq. W- 17)

Where:
Ea,i = Annual GHG i (either CH4 or CO2) volumetric emissions at ambient conditions
MT = meter volumetric reading of gas emissions per unit time, under ambient conditions
T = total time the compressor associated with the venting s operational in the reporting year
Mi = mole percent of GHG i in the vent gas; use the appropriate gas composition in paragraph
(u)(2) of this section
Determination of gas emission rate (MT)
If the rod packing case is connected to an open ended vent line then use one of the following two
methods to calculate emissions.
(i) measure emissions a from all vents (including emissions manifolded to common vents)
methods set forth in 98.234 (c).
(ii) use a temporary meter such as, but not limited to, a vane anemometer, or a permanent meter
emissions manifolded to a common vent) including rod packing vents, unit isolation valves, and
blowdown valves according to methods set forth in 98.234(b). If the rod packing case is not
equipped with a vent line, use the following method to estimate emissions: (i) You must use the
1525
methods described in 98.234(a) to conduct annual leak detection of fugitive emissions from the
packing case into the distance piece, or from the compressor crank case breather cap or vent with
a closed distance piece.
(ii) Measure emissions using a high flow sampler, or calibrated bag, or appropriate
meter according to methods set forth in 98.234(d).
Measurement frequency (MT)
Conduct one measurement for each compressor in each of the operating modes that
occurs during a reporting period:
(i) Operating
(ii) standby pressurized
(iii) not operating, depressurized.
Step 2: Calculate CH4 and CO2 volumetric emissions at standard conditions using calculations in
paragraph (t) of this section
Step3: Estimate CH4 and CO2 volumetric emissions and mass emissions from volumetric
natural gas emissions using the calculations in paragraphs (u) and (v) of this section.
WCI Recommendations: WCI recommends adopting the USEPA methodologies for
reciprocating and centrifugal compressors with no modifications. Both these methods utilize
direct measurement of emissions which will provide accurate, cap-and-trade quality emissions
quantification for compressors in gas processing plants.
Response: EPA has revised todays final rule for compressor monitoring. Please see Section
II.F of the preamble to todays final rule and Compressor Modes and Threshold (EPA-HQ
OAR-2009-0023).
Section 98.233(o) Centrifugal compressor wet seal degassing vents
Section 98.233(o) does not currently prescribe that wet seal vents be surveyed in each of the
operational modes that occur during a reporting period (as does for reciprocating compressor
vents). But the TSD 221 does acknowledge these emissions differ in various operating modes.
Therefore, it is not clear if it was the EPAs intent to have the three mode requirement apply to
centrifugal compressors. El Paso requests clarification on the requirement to survey in various
modes of operation. (Note that because centrifugal compressors with wet seals cannot remain
pressurized when offline, only two modes of operation are possible: operating and not operating,
depressurized.) If this requirement was intended to apply, El Paso recommends that centrifugal
compressors with wet seals be surveyed as per the As Found approach described under Section
XV and that emission factors be calculated as presented in Table 7 below. Table 7 presents a
sample calculation of the annual leak rate from a wet seal centrifugal compressor unit that is
blown down when off line. In this example, hypothetical emission factors are used to represent
221
See Appendix M of the Greenhouse Gas Emissions Reporting from the Petroleum and Natural Gas Industry,
Background Technical Support Document, page 160.
1526
the key sources. This example also assumes that the compressor unit in question has one
compressor wet seal, one blow down valve, one pressure relief valve, and one pair of unit
isolation valves. Finally, the compressor unit is assumed to be Operating 75% of the year
(6570 hours) and Not Operating, Depressurized the remaining 25% of the year (2190 hours).
Therefore, as shown in Table 7, to calculate the contribution of wet seal leakage while the
compressor unit is operating, the hypothetical emission factor for Operating wet seals (120.0
scfh CH4) is multiplied by the number of wet seals (1) and again multiplied by the activity factor
(6570 hours per year that the compressor is in Operating mode). This result was then divided
by 1000 to convert to Mcf/yr (Mcf = 1000 cubic feet in US natural gas industry terminology)
resulting in 788 Mcf CH4/year. Similar calculations are made for each of the other component
categories, with the results from each category being totaled to calculate the overall annual
compressor unit leak rate. Note that in practice, emissions will be calculated by using a
combination of unit specific measurement data and operating company emission factors. Actual
measurement data will be used for each unit as available, and the emission factors will be used to
project leakage from the mode which was not measured. This will make the calculations more
facility specific.
Table 7: Sample Calculation of Emissions from a Centrifugal Compressor Unit With Wet (Oil)
Seal (Basis: 75% Utilization)
Response: EPA has clarified the modes in which compressors must be measured and
corresponding methodologies in todays final rule. For more information please see Section II.F
of the preamble to todays final rule and Compressor Modes and Threshold (EPA-HQ-OAR
2009-0023).
1527

Proposed Rule Section 98.233(o) Centrifugal compressor wet seal degassing vents.El Paso
requests clarification on the operating mode requirements for surveys of centrifugal wet seal
vents. It is our interpretation that Section 98.233(o) does not currently prescribe that wet seal
vents be surveyed in each of the operational modes that occur during a reporting period.
Response: EPA has clarified the modes in which compressors must be measured and
corresponding methodologies in todays final rule. For more information please see Section II.F
of the preamble to todays final rule and Compressor Modes and Threshold (EPA-HQ-OAR
2009-0023).

Centrifugal Compressor Wet Seal Degassing Vents 98.233(o) defines emission estimates for
centrifugal compressor wet seal degassing vents based on annual measurement of the wet seal oil
degassing tank vent. In general, INGAA accepts the proposed vent measurement requirements
for wet seal degassing, with the request that the following three issues be addressed in the Final
Rule:
As discussed in Comment VI, operators should be allowed to elect to conduct direct
measurement of common vent lines, where applicable, and report the actual annual volume
measured;
In addition to flow measurement meters prescribed in the Proposed Rule, alternative tools for
vent gas measurement should be allowed, provided that the vented rate is within an acceptable
range of the instrument. These alternatives should include calibrated bags (98.234(c)) and highvolume samplers (98.234(d)); and
Engineering units should be provided for parameters in equation W-16 and these units should
be consistent with the referenced calculations in 98.233(t) and 98.233(v). This issue is
highlighted here, and a similar general statement is provided in Comment XIII regarding review
and correction of equations, especially broadly-applied equations such as those in 98.233(t) and
(v).
Furthermore, as discussed in Comment X, reporting individual compressor throughput is not
feasible or practical and should be deleted from the Rule
Response: EPA disagrees with this comment that centrifugal compressor venting be allowed to
only conduct direct measurement of common manifolded vent lines. The rule allows installation
of a port on each vent line to the manifold, for insertion of a temporary meter, and these ports
can be installed at ground level. Typically, centrifugal compressor venting emissions vary with
1528
the mode of operation of the compressor. Generally, the emissions are highest when the
compressor is operating and lower when they are in the not operating depressurized mode.
However, when the compressor is not operating depressurized, there may be leakage of natural
gas through the unit isolation valve, particularly if the valve seat has become fouled and will not
completely close. Hence to correctly characterize annual emissions from centrifugal
compressors, estimation of emissions in two compressor modes, operating, and not operating
depressurized is required. A temporary meter such as vane anemometer or permanent meter
such as orifice meter can be used to measure emissions from the vents. EPA disagrees with
allowing measurement of seal oil degassing vent emissions with calibrated bags or high-volume
samplers, because the volume of gas from the degassing tank vent is typically well in excess of
the capacity of these measurement devices. EPA agrees with the comment to add engineering
units to the equation for centrifugal compressors, and has revised todays final rule.
Commenters made several assumptions which were not consistent with the proposed rules
intent. One was the assumption that each compressor would need a meter to determine
compressor throughput. For more information, please see response to comment EPA-HQ-OAR
2009-0923-1206-63.

Section 98.233(o) Centrifugal compressor wet seal degassing vents. The variable Mi in Equation
W-16 should be a volume percent of GHG to make the units work out.
Response: EPA disagrees with this comment to change the variable Mi in this equation to
volume percent. The term mole percent has no units and neither does the term volume
percent. In either case, it would not be expected to make any difference to the units in the
equation. If we consider the GHGs as near-ideal gases, then under that condition, the mole
fraction (which is the same as mole percent) would be identical with volume percent (or volume
fraction). Hence, no change is required for the variable Mi to make it volume percent.

Aka recommends that in Section 98.233(o) the rule should allow for the use of API
Compendium factors to estimate emissions from centrifugal compressor wet seal degassing
vents.
Response: EPA disagrees with this comment. EPA experience, new data and increased
knowledge of industry operations and practices have highlighted the fact that centrifugal wet seal
1529
degassing venting emissions estimates based on the factors used in API Compendium are
outdated and potentially understated. For a discussion please see the Technical Support
Document EPA-HQ-OAR-2009-0923. Further, the API compendium factors do not account for
venting emissions from unit isolation valves. EPA determined that direct measurement of
centrifugal wet seal degassing venting is required to maintain the necessary quality of data to
inform policy.

Centrifugal compressor wet seal degassing vents
Section 98.233(o): For the same reasons cited above for reciprocating compressor rod packing
venting (excessive cost, well beyond what EPA originally estimated), direct measurement is not
appropriate for centrifugal compressor wet seal degassing vents. Therefore, IPAMS requests that
the final rule allow for the use of API Compendium factors to estimate emissions from
centrifugal compressor wet seal degassing vents.
Response: EPA disagrees with this comment. Please see response to comment EPA-HQ-OAR
2009-0923-1080-24.

Section 98.233(o) Centrifugal compressor wet seal degassing venting.
API requests that EPA broaden the available methodologies for determining flow from
centrifugal compressor wet seal degassing vents to include a tracer gas methodology. A
description of this methodology and example follow:
1. Introduce an inert tracer gas (such as N2) into the atmospheric degassing tank at a measured
(e.g. rotometer) or known (e.g. critical orifice) rate. The rate must be high enough to yield a
tracer gas concentration well above the analysis methodology detection limit.
2. Allow sufficient time for the tracer gas to become well mixed in the degassing tank with the
gas evolving from the seal oil.
3. Sample the vent stream from the degassing tank and analyze the gas using the appropriate
analysis technique (e.g. Gas Chromatography).
4. Calculate the CH4 and CO2 content of the vent gas based on the ratio of CO2 or CH4 to tracer
gas.
5. Multiply this ratio times the tracer gas flow rate to determine the flow rate of CO2 and CH4
from the vent. If the vent is routed to a flare or other combustion device, also calculate the flow
of all hydrocarbon species in the vent gas analysis which will yield CO2 when combusted.
6. Use the appropriate techniques in Section 98.233 to convert these flows into GHG terms.
1530
Response: EPA disagrees with this comment. Ensuring a good degree of mixing of the tracer
gas with the vented gas to ensure representative sampling for analysis would be of concern.
Another potential concern might be of an altering of the solution thermodynamics such that the
quantity and quality of the venting gas might be affected by introducing the inert gas. EPA
determined that direct measurement of centrifugal wet seal degassing venting is required to
maintain the necessary quality of data to inform policy.

Section 98.233(o) Centrifugal compressor wet seal degassing venting. BP requests that EPA
broaden the available methodologies for determining flow from centrifugal compressor wet seal
degassing vents to include a tracer gas methodology. A description of this methodology and
example follow:
1. Introduce an inert tracer gas (such as N2) into the atmospheric degassing tank at a measured
(e.g. rotometer) or known (e.g. critical orifice) rate. The rate must be high enough to yield a
tracer gas concentration well above the analysis methodology detection limit.
2. Allow sufficient time for the tracer gas to become well mixed in the degassing tank with the
gas evolving from the seal oil.
3. Sample the vent stream from the degassing tank and analyze the gas using the appropriate
analysis technique (e.g. Gas Chromatography).
4. Calculate the methane and CO2 content of the vent gas based on the ratio of CO2 or CH4 to
tracer gas.
5. Multiply this ratio times the tracer gas flow rate to determine the flow rate of CO2 and CH4
from the vent.
a. If the vent is routed to a flare or other combustion device, also calculate the flow of all
hydrocarbon species in the vent gas analysis which will yield CO2 when combusted.
6. Use the appropriate techniques in 98.233 to convert these flows into GHG terms. BP also
requests that a single representative volume sample estimate be allowed for same manufactured
compressors in the same service and operating parameters. This reduces the data collection
burden while gathering adequate data to estimate emissions.
Example:
[Table]
1531
Response: EPA disagrees with this comment. Please see response to comment EPA-HQ-OAR
2009-0923-1151-124.
13.2.7.2
RECIPROCATING COMPRESSORS

Operational Modes of Compressor Units
Although the mode of a compressor unit can substantially affect the leakage for individual units,
the average differences may not be as great as indicated in the EPA source document mentioned
in the technical support document for the proposed rule. This Star program Lessons Learned
publication (EPA, 2004) states that rod packings on idle units may leak at four times the rate of
rod packings on running units.
1532
However, the primary references for this document are personal communications and better data
is currently available. For instance, the original GRI survey of compressor stations (Indaco,
1995) found a 19% increase in idle versus running rod packings, while in the PRCI study
(Howard et al., 1999 cosponsored by PRCI, GRI, and EPA), idle rod packings were found to
leak on average at a 46% greater rate than running rod packings.
A more extensive data set has been compiled by El Paso Natural Gas Company. This data was
primarily collected by outside contractors and has approximately ten times as many
measurements as the PRCI data set. This data indicates an average of only 6% increase in
leakage from idle seals compared to running seals. Consequently, the difference in leakage
between running units and idle but pressurized units is probably not as significant as indicated in
the EPA Lessons Learned source.
For the depressurized units, the EPA Lessons Learned document suggests an average leak rate of
23.3 scfm. This is in line with the original GRI study of 26.1 scfm. However, the PRCI study
indicated a leak rate of 6.78 scfm (although this was based on a smaller data set than GRI it was
a more diverse set of stations). Finally, El Pasos data set (which has over four times more data
points for unit valves than the GRI study) indicates a leak rate of 9.21+/-2.64scfm.
To estimate the differences between modes, lets assume a model of a reciprocating compressor
unit that has one blow down valve, four compressor cylinders and therefore four rod packing
seals, and one pair of unit isolation valves. When either running or in standby pressurized mode,
there is potential leakage across the blow down valve and through the rod packings. When
depressurized, the blow down valve is open and leakage measured from that vent is due to
leakage across the unit isolation valves.
Table A-1 summarizes the available emissions factors and the results of the model using the
available data.
Table A-1. Emission Factors and Projected Leak Rates from Related Leak Studies Component
EPA
1533
Based on these projections, we might expect idle/pressurized units to leak on average between
5% and 43% more than running units. Depressurized units might leak on average slightly less
than running units to as much as four times greater.
If the El Paso and PRCI data are representative, then there is little difference between operating
modes on average, although for individual units there may be substantial differences. If this is
the case, sampling units in an As Found mode should provide an excellent representation of
total leakage.
If the EPA estimates (or the original GRI factors for unit isolation valves) are more
representative, then the depressurized mode may leak on average three to four times greater than
the running mode. If this is the case, then it will be important to have a representative sampling
of units that spend the majority of their time depressurized.
The three year window recommended under Suggestion 4 in the body of this letter should
address this situation. By requiring at least 50% of units to be surveyed in their primary mode
over a three year period, units can be surveyed As Found annually, which will probably
provide a representative sample of each mode. However, the requirement to survey at least 50%
of units in their primary mode over a three year period provides a safety net to ensure
representative sampling while still allowing the flexibility to avoid disruptions in gas
transportation and unnecessary blow downs of units.
Response: EPA agrees that compressor venting be measured in the mode as found, and has
clarified this in todays final rule and allows reporters to schedule the annual measurement of
each compressor, and to measure the compressor in the mode as it exists at the time the annual
measurement is taken. EPA considered comments received and determined that reporter based
emission factors can be developed and applied to the reporters other compressors that were not
measured in that mode. Therefore to lessen undue burden, EPA is requiring the development of
reporter based emission factors from these measurements that reporters must apply appropriately
to all compressors for the total time each compressor is operated in each mode. However each
compressor must be measured at least once every three years in the not operating and
depressurized mode without blind flanges in place. Based on comments received compressors
come down for maintenance during a three year time period. EPA considered comments
received that indicate measuring emissions in the not operating and depressurized mode would
be difficult as this mode may not occur frequently. EPA also considered that in the not operating
and depressurized mode the potential emissions from the unit isolation valve would be unique to
each compressor and the ability of each valve to close completely. Therefore, to lessen burden,
reporters are allowed to schedule the measurement of not operating and depressurized mode at
any time during those three consecutive years. For more information please see Compressor
Modes and Threshold memo under rulemaking docket EPA-HQ-OAR-2009-0923. Regarding
the comment on studies by GRI (Indaco 1995), PRCI (Howard et al., 1999 by PRCI, GRI, and
EPA), and El Paso Natural Gas Company, the supporting data of these studies are not known to
be publicly available.
1534

Reciprocating compressor rod packing venting
Calculate annual CH4 and CO2 emissions from each reciprocating compressor rod packing
venting as follows:
Step 1: Calculate CH4 and CO2 volumetric emissions at ambient conditions as follows:
E [subscript a, i] = MT * T * M [subscript i] (Eq. W- 17)
Where:
E [subscript a, i] = Annual GHG i (either CH4 or CO2) volumetric emissions at ambient
conditions
MT = meter volumetric reading of gas emissions per unit time, under ambient conditions
T = total time the compressor associated with the venting s operational in the reporting year
M [subscript i] = mole percent of GHG i in the vent gas; use the appropriate gas composition in
paragraph (u)(2) of this section
Determination of gas emission rate (MT)
If the rod packing case is connected to an open ended vent line then use one of the following two
methods to calculate emissions.
(i) measure emissions a from all vents (including emissions manifolded to common vents)
methods set forth in SECTION 98.234 (c).
(ii) use a temporary meter such as, but not limited to, a vane anemometer, or a permanent meter
emissions manifolded to a common vent) including rod packing vents, unit isolation valves, and
blowdown valves according to methods set forth in SECTION 98.234(b).
If the rod packing case is not equipped with a vent line, use the following method to estimate
emissions:
(i) You must use the methods described in SECTION 98.234(a) to conduct annual leak detection
of fugitive emissions from the packing case into the distance piece, or from the compressor crank
case breather cap or vent with a closed distance piece.
(ii) Measure emissions using a high flow sampler, or calibrated bag, or appropriate meter
according to methods set forth in 98.234(d).
1535
Measurement frequency (MT)

Conduct one measurement for each compressor in each of the operating modes that occurs
during a reporting period:
(i) Operating
(ii) standby pressurized
(iii) not operating, depressurized.
Step 2: Calculate CH4 and CO2 volumetric emissions at standard conditions using calculations in
paragraph (t) of this section
Step3: Estimate CH4 and CO2 volumetric emissions and mass emissions from volumetric natural
gas emissions using the calculations in paragraphs (u) and (v) of this section.
WCI Recommendations: WCI recommends adopting the USEPA methodologies for
reciprocating and centrifugal compressors with no modifications. Both these methods utilize
direct measurement of emissions which will provide accurate, cap-and-trade quality emissions
quantification for compressors in gas processing plants.
Response: EPA has reviewed this comment and has retained direct measurement of compressor
venting in todays final rule for natural gas processing.

Organization: Verdeo Group
Commenter: John Savage
S98.233 (p) Reciprocating compressor rod packing venting
Section (2) stipulates the use of bagging technique or a temporary meter for measurement of
emissions If the rod packing case is connected to an open ended vent line.
Section (3) stipulates the use of a high volume sampler for measurement of emissions If the rod
packing case is not equipped with a vent line..
Because high-volume samplers are particularly effective at measuring low volume methane
emissions from OELs, Verdeo recommends that such high-volume samplers be included as a
measurement option under Section (2).
This would be consistent with best practices as advocated by the EPAs Gas STAR program and
would also provide a more convenient and less capital intensive option for many reporting
companies.
1536
Response: EPA agrees with the comment to include the option to use a high volume sampler to
measure emissions from reciprocating compressor venting, and has revised todays final rule.

(98.233 (p)) Can we assume that if our reciprocating compressor rod emissions are captured by
vapor recovery that they do not have to be calculated or reported.
Response: EPA agrees that if reciprocating compressor venting are captured by vapor recovery
or routed to a flare, they do not need to be reported. If the vapor recovery unit is not in
operation, companies must follow methodologies for measuring reciprocating and centrifugal
compressor emissions. Please see 98.233 (p) of the rule for more information.

Proposed Rule Section 98.233(p) Reciprocating compressor rod packing venting.(KEY
ISSUE) El Paso requests the EPA to allow surveys of compressors to take place in an as found
mode. While our overall principle recommendation is the same as INGAAs, we have provided
options and justified our recommendations based on actual field experience at El Paso sites and
review of existing literature on this subject.
Response: EPA agrees to allow compressor venting measurement in the as found mode.
Please see response to comment EPA-HQ-OAR-2009-0923-0055-16.

Section 98.233(p) Reciprocating compressor rod packing venting: El Paso recommends
monitoring in As Found mode. (Note: This section also applies to other vents as well).
Section 98.233(p) requires that all compressor vents be surveyed in each of the operational
modes that occur during a reporting period:
1. Operating
2. Standby pressurized
1537
3. Not operating, depressurized

El Paso recognizes and mentioned in its comments 222 to the EPA proposed reporting rule last
year, that the operating mode of compressor units impacts leakage from key compressor
components such as rod packing seals, blow down valves, and unit isolation valves. However, a
survey of all three modes as required by the EPA in the Subpart W proposal creates enormous
challenges and will result in greenhouse gas emissions (GHG) that otherwise, would not have
occurred under normal operations.
To alleviate these problems, as the first step, the EPA should take into account that compressor
units will primarily fall under one of the following operating scenarios:
The proposed monitoring requirements do not consider challenges associated with atypical
modes of operation. Currently, Subpart W does not take into account that the following modes of
operation are considered atypical and probably occur less than 7% of the time (see Appendix I):
1. Scenario 1 - For units that are normally blown down when off line, only a very short amount
of time is spent in the standby, pressurized mode prior to being blown down (not operating,
depressurized mode).
2. Scenario 2 - Units that are normally in the standby, pressurized mode when off line will
only be depressurized for maintenance or during emergency conditions.
Additionally, it does not consider that requiring measurement of both typical and atypical modes
of operation would result in unintended GHG emissions from unit blowdowns, disruption of
operations and supply, and non-representative conditions. Furthermore, monitoring compressor
units in atypical modes has the additional challenge of tracking the operating time in these
modes, which is not a common practice by most companies today, and would be a major
undertaking not justified by the level of improvement in capturing emissions. The increase in
GHG emissions, difficulty in obtaining representative sampling, and supply disruption problems
due to the three-mode sampling requirement are important aspects of the problem, which EPA
appears not to have considered in proposing the three-mode requirement. EPA should consider
the alternatives El Paso suggests below to help alleviate these problems.
To achieve the EPAs goal of collecting meaningful and representative emissions data, the
abovementioned operating scenarios should be the focal point of the EPAs monitoring proposal
while considering the as found mode of operation of a given unit. The as found mode of a
compressor is the mode in which it is operating on the day of a survey. Although El Paso
222
Docket ID No EPAHQOAR20080508.
1538
compressor data shows that the as found approach will result in a large sample size of all
modes 223 that can be related to the annual emissions from the facility, El Paso recommends a
hybrid monitoring approach to ensure an ample data set of all modes over a reasonable amount
of time:
1. Each unit located at an applicable facility shall be surveyed once a year in the as found
mode; and
2. A minimum of 33% of the total population of units that spend more than 10% 224,225,226 . The
not operating, depressurized mode shall be based on the operation of the unit during the year
prior to the 3 year period.
Note that the 3 year time period referenced above will not preclude the submission of emissions
data in the not operating, depressurized mode during the first reporting year as discussed in the
appendix.
Again, it is anticipated that the as found approach will result in a large sample size of all
modes. However, for any given unit it would be acceptable if it were surveyed in its as found
mode all three times during the three year period.
In practice, each compressor unit would be surveyed in the mode in which it was found when the
survey took place. This data would then be used to calculate emissions factors for each key
component category in each mode. Total unit emissions would then be calculated as follows:
1. For the mode in which the unit was surveyed (as found), the emissions will be based on
actual measurements.
2. For the other mode, the emissions will be based on the number of key components, and
emission factors developed for this mode based on company-wide sampling results in this mode.
3. The atypical mode will be rolled up with one of the typical modes as described in Equation
4a or 4b in Appendix I.
Details regarding the development of emission factors, including sample calculations, are
223
Over the past three years (2007 2009), compressor units in the El Paso transmission system spent over 50% of
their time in each of the modes as follows: Operating (28.6%), Standby, Pressurized (34.7%), and Not Operating,
Depressurized (36.8%). Consequently, As Found sampling should provide a representative sample of each
mode.
224
The basis for including only units that are in the "not operating, depressurized" mode over 10% of the time is that
emissions from this mode will no longer be dominant at this point. Based on EPA's estimates from the Lessons
Learned document in the TSD, the "not operating, depressurized" mode might emit on average 2.7 times more
than the "Operating" mode for a unit with four compressor seals. If this is true, for units that are depressurized less
than 10% of the time, the depressurized mode contributes less than 25% of the total emissions.
225
Of their time in not operating, depressurized mode, shall be surveyed in that mode once every 3 years.
226
The three year window is required because it will not be possible to bring many units offline until their normal
maintenance cycle arises.
1539
presented in Appendix I. The methodology is illustrated in the table below.
Lastly, clarification is needed between Section 98.233(p)(2)(i) which requires the measurement
of compressor blow down vents along with other compressor unit vents and Section 98.233(q)(3)
which indicates that compressor blow down vents should be surveyed using an IR camera and
then the leaker emission factor applied to calculate leakage.
Please see response to comment EPA-HQ-OAR-2009-0923-0055-16. In regard to blowdown
venting, the commenter has misinterpreted the proposed rules intent. Todays final rule requires
compressor blowdown venting, or through valve leakage, to be reported under Sections
98.233(o) and (p). Todays final rule also requires equipment leak emissions from compressor
blowdown valves to be reported under Section 98.233(q).

Support Documentation for As Found Monitoring Proposal
Monitoring Challenges
1. Surveying all three modes will require an additional blow down event per unit, because if a
unit is on line it would have to be taken off line and then blown down to survey the
depressurized mode. This would result in an additional blowdown event that would increase
GHG emissions. On the other hand, if a unit were already blown down, it would have to be
pressurized and put on line, and then after the measurements were made, it would be taken back
off line and blown down. This would also increase GHG emissions. Hence, the monitoring
requirement will result in increased greenhouse gas emissions that otherwise would not have
occurred under normal operations as per the case shown below:
At least 50% of compressor units in El Pasos transmission system are currently depressurized
when off line and as a result, already blown down. So all of these units would have to be
pressurized and put on line to make a measurement before being put back off line and blow
down a second time. All units found operating during the survey would have to be taken off line
in order to measure either the standby pressurized mode and the not operating
depressurized mode.
1540
2. Putting a unit on line temporarily may not provide a representative measurement since the rod
packings may not be properly warmed up or it may have to be placed in a re-circulating mode 227
if there is no extra gas capacity in the pipeline.
3. It may not be feasible to take a unit off line due to pipeline demand requiring a unit to come
off line might disrupt gas supply and operations. This may be especially true in the production
sector or at transmission stations with only centrifugal units where there may not be any
additional units with which to switch.
4. Generally, companies in the oil and gas sector keep track only of the compressor operating
hours, so that the only current reliable information is whether a compressor unit is operating or
off-line. It is El Pasos understanding that there is no current standardized industry practice or
method of determining whether an off-line compressor is pressurized or depressurized. Usually,
standard operating procedures are employed to estimate the hours when the unit is off-line. If the
standard procedure was to blow down the unit when off- line, then it would be assumed that all
off-line hours were spent depressurized. Similarly, if the standard procedure was to leave the unit
pressurized when off-line, it would be assumed that all off-line hours were pressurized.
Modifying the compressors Remote Terminal Units (RTUs) for all affected units to track hours
in atypical modes would be a major undertaking not justified by the level of improvement in
capturing emissions.
5. Units that are usually depressurized when off line might spend a very short amount of time
pressurized prior to being blown down, and units that are left pressurized while off line might be
depressurized for short periods for maintenance. However, these are atypical conditions. For
instance, in 2009, compressor units in El Pasos transmission system that are typically left in
standby pressurized mode when off-line were operating 31.6% of the time, in standby
pressurized mode 61.9% of the time, and unavailable 6.5% of the time. This unavailable
designation comes from El Paso RTU data and the unit may or may not be depressurized during
this time, so in the worst case the not operating depressurized mode (the atypical mode for
units typically left pressurized when off line) only occurs for 6.5% of the time.
For units which are typically depressurized when off-line, detailed data on the amount of time
spent in standby pressurized (the atypical condition) is not available. However, in most cases
these units are only left in the standby pressurized for a short period (less than one hour) prior
to being depressurized.
Effect of Operational Mode on Leakage
Although the mode of a compressor unit can substantially affect the leakage for individual units,
the average differences may not be as great as indicated in the EPA source document mentioned
227
In a recirculating mode, the gas leaving the discharge side of the compressor flows through a bypass valve and
into the suction side of the compressor. As a result, suction and discharge pressures are essentially equal in
recirculating mode. During typical compressor operations, the difference between suction and discharge pressures
are at least 200 psig.
1541
in the technical support document for the proposed rule. This Star program Lessons Learned
publication (EPA, 2004) 228 states that rod packings on idle units may leak at four times the rate
of rod packings on running units.
However, the primary references for this document are personal communications and better data
is currently available. For instance, the original GRI survey of compressor stations (Indaco,
1995) 229 found a 19% increase in idle versus running rod packings, while in the PRCI study
(Howard et al., 1999 cosponsored by PRCI, GRI, and EPA), idle rod packings were found to
leak on average at a 46% greater rate than running rod packings.
A more extensive data set has been compiled by El Paso. This data was primarily collected by
outside contractors and has approximately ten times as many measurements as the PRCI data set.
This data indicates an average of only 6% increase in leakage from seals in idle pressurized
mode compared to seals in running mode. Consequently, the difference in leakage between
running units and idle but pressurized units is probably not as significant as indicated in the EPA
Lessons Learned source.
For the depressurized units, the EPA Lessons Learned document suggests an average leak rate of
23.3 scfm. This is in line with the original GRI study of 26.1 scfm. However, the PRCI study
indicated a leak rate of 6.78 scfm (although this was based on a smaller data set than GRI it was
a more diverse set of stations). Finally, El Pasos data set (which has over four times more data
points for unit valves than the GRI study) indicates a leak rate of 9.21 +/- 2.64 scfm.
To estimate the differences between modes, lets assume a model of a reciprocating compressor
unit that has one blow down valve, four compressor cylinders and therefore four rod packing
seals, and one pair of unit isolation valves. When either running or in standby pressurized mode,
there is potential leakage across the blow down valve and through the rod packings. When
depressurized, the blow down valve is open and leakage measured from that vent is due to
leakage across the unit isolation valves.
Table 1 summarizes the available emissions factors and the results of the model using the
available data.
Table 1: Emission Factors and Projected Leak Rates from Related Leak Studies
228
EPA, 2004. Reducing Emissions when Taking Compressors Off-Line. EPA430-B-04-001, February 2004.
229
Indaco, 1995. Leak Rate Measurements at US Natural Gas Transmission Compressor Stations. GRI-94/0257.37.
Gas Research Institute, 8600 West Bryn Mawr Avenue, Chicago, IL 60631.
1542
Based on these projections, we might expect idle/pressurized units to leak on average between
5% and 43% more than running units. Depressurized units might leak on average slightly less
than running units to as much as four times greater.
The El Paso and PRCI data indicate there is little difference between operating modes on
average, although for individual units there may be substantial differences. If this is the case,
sampling units in an As Found mode should provide an excellent representation of total
leakage.
Calculation of System-wide Emissions Using Emission Factors
To aid in understanding the relative contributions of various leak sources to total emissions, a
simple model of compressor station fugitive methane emissions can be constructed based on the
major sources of leakage. This model is illustrated in Figures 1, 2, and 3.
Figure 1 shows the leak sources for a unit in the Operating mode. In this mode leakage may
occur through the rod packing seals and across the blow down valve and pressure relief valve,
which are closed when the unit is operating. Much smaller leakage may also occur at standard
components (such as connectors and valves) on the compressors themselves but this leakage is
addressed by Section 98.233(q) of Subpart W. For this case, the total leakage from the key
vented components while in the operating mode is calculated per Equation 1:
1543
For most compressor units, the number of blow down and pressure relief valves will be one of
each, but there are some units that have more. Additionally, the emission factors for both blow
down valves and pressure relief valves are not expected to differ between units that are in the
Operating mode and in the Standby Pressurized mode, i.e., these factors are not impacted by
mode.
Figure 2 shows the leakage for a unit in the Standby Pressurized mode. The leakage in this
mode occurs at the same sources as in the Operating mode except that the leakage through the
rod packing seals may differ between the two and consequently is represented by a different
emission factor. For this case, the total leakage from the key vented components while in the
standby pressurized mode is calculated per Equation 2:
Figure 3 represents the leakage from a reciprocating compressor unit that is in the Not
Operating Depressurized mode. When a unit is depressurized (blown down), the only source
of leakage will be through the unit isolation valves, which are closed in order to isolate the unit
from the main pipeline. Leakage through these valves travels through the open blow down valve
and is measured at the blow down vent. The leakage in this case is calculated pursuant to
Equation 3:
1544
The unit isolation valves are counted as a pair because there will always be at least one suction
valve and one discharge valve and measurements at the blow down vent cannot distinguish
which valve might be leaking. Typically there will be only one pair per unit at transmission
compressor stations; however, storage facilities may have more than one pair per unit.
Annual leakage from a compressor unit can then be calculated using known activity data
(amount of time it spends in two of the three modes Operating, Standby Pressurized, Not
OperatingDepressurized) and the leak rate in two of the three modes per Equation 4a or 4b:
Figure 1: Leakage from Reciprocating Compressor Unit in Operating Mode
1545
Figure 2: Leakage from Reciprocating Compressor Units in Standby Pressurized Mode
Figure 3: Leakage from Reciprocating Compressor Units in Not Operating--Depressurized

Mode
1546
Example Calculations of Leak Rates Using Emission Factors

Table 2 presents a sample calculation of the annual leak rate from a reciprocating compressor
unit that is left pressurized when taken off line. In this example, hypothetical emission factors are
used to represent the key sources. This example also assumes that the compressor unit in
question has four compressor rod packing seals, one blow down valve, one pressure relief valve,
and one pair of unit isolation valves. Finally, the compressor unit is assumed to be Operating
75% of the year (6570 hours) and Not OperatingDepressurized the remaining 25% of the
year (2190 hours).
Therefore, as shown in Table 2, to calculate the contribution of rod packing seal leakage while
the compressor unit is operating, the hypothetical emission factor for Operating rod packing
seals (99.0 scfh CH4) is multiplied by the number of rod packing seals (4) and again multiplied
by the activity factor (6570 hours per year that the compressor is in Operating mode). This
result was then divided by 1000 to convert to Mcf/yr (Mcf = 1000 cubic feet in US natural gas
industry terminology) resulting in 2602 Mcf CH4/year. Similar calculations are made for each of
the other component categories, with the results from each category being totaled to calculate the
overall annual compressor unit leak rate.
Note that in practice, emissions will be calculated by using a combination of unit specific
measurement data and operating company emission factors. Actual measurement data will be
used for each unit as available, and the emission factors will be used to project leakage from the
mode which was not measured. This will make the calculations more facility specific.
Table 3 presents a similar example for a reciprocating unit that is depressurized when off-line.
1547
Table 2: Sample Calculation of Emissions from a Reciprocating Compressor Unit --Pressurized

When Not Operating (Basis: 75% Utilization)
Table 3: Sample Calculation of Emissions from a Reciprocating Compressor Unit that is

Depressurized When Not Operating, (Basis: 75% Utilization)
subpart W. Please see Section II.F of the preamble to todays final rule. EPA agrees to allow
compressor venting measurement in the as found mode with the exception of measuring
emissions from compressors in standby, depressurized at least once every three years. Please see
Commenter: Kim Dang
1548
Reciprocating Compressor Rod Packing Venting. Kinder Morgan believes that it is impractical,
counterproductive, and unnecessary to require annual emissions measurement for all three
operating modes (operating, standby pressurized, and depressurized) of each reciprocating
compressor. 230 This requirement will result in increased and unnecessary methane emissions
when compressors are removed from service to take measurements in depressurized mode, as
an additional blowdown will be required that would not have otherwise taken place. Moreover,
in order to fulfill this requirement, pipeline owners will likely force the compressor into all three
operating modes while the measurement technicians are at the facility. The requirement to
measure in all three operating modes will result in unnecessary stress on the equipment.
Requiring monitoring at each of the three operating modes would also reduce gas transportation
volumes and could result in disruption of service to industrial and residential customers.
Although the requirement will result in the collection of more data, there is no reason to believe
that EPA would not collect significant quantities of data on compressors in all three operating
modes even in the absence of the requirement.
Thus, Kinder Morgan recommends that emissions measurements be taken in whatever mode the
reciprocating compressor is in at the time of the measurement, rather than for all three operating
modes at each individual compressor. Given the large overall number of compressors that will be
monitored under Subpart W throughout the industry, this modification would still provide EPA
with abundant data concerning the emissions profile of compressors in each operating mode. At
the same time, the modification would minimize operational disruptions at pipelines and avoid
the requirement to deliberately cause otherwise unnecessary GHG emitting blowdown events.
Response: EPA agrees to allow compressor venting measurement in the as found mode with
the exception of measuring emissions from compressors in standby, depressurized at least once
every three years. Please see response to comment EPA-HQ-OAR-2009-0923-0055-16.

Commenter: Kim Dang
EPA should not require that reciprocating compressors be monitored in all three operating modes
because this is impractical and will result in excess emissions, but should instead require that
reciprocating compressors be monitored as found.
230
See Proposed 40 C.F.R. SECTION 98.233(p)(4).

1549

The rule would require that measurements of emission from reciprocating compressor rod
packing vents be conducted in three operating modes - operating, not operating/pressurized, and
depressurized - but only one of these measurements is ultimately used in the emissions
calculation. Taking these measurements will result in an increase in GHG emissions as the units
are vented to obtain the measurements. The rule also does not indicate a length of time or
operating mode when metering emissions from wet seals. Dominion requests that these
measurements be taken in either "typical" operating mode or "as found" mode and that
monitoring times be specified.

Reciprocating Compressor Rod Packing Annual Vent Measurements Add Complexity and
Increase GHG Emissions. Measurement is Reasonable when Access is Available, but Safety
Concerns and Alternatives to Three-Mode Testing Must be Addressed
98.233(p) identifies requirements for estimating emissions from reciprocating compressor rodpacking vents that include annual three-mode testing (i.e., measurements during each
operational mode that occurs during a reporting period): operating; standby pressurized; and not
operating, depressurized. INGAA is receptive to conducting annual reciprocating compressor
rod-packing vent measurements at applicable facilities. However, it is imperative that EPA
reconsider mode testing requirements, and thus rule refinements are required. INGAA strongly
recommends that annual tests be conducted at a single as-found operating mode each year.
Vent access safety concerns, manifolded vent lines, and the creation of GHG emissions from unit
blowdowns for the sole purpose of depressurized-mode testing present significant obstacles that
must be practically resolved.
A. Annual Three-Mode Testing is Complicated and Would Increase GHG Emissions
As required in 98.233(p)(4), three-mode testing introduces significant logistical complexity,
burden, cost, and results in unnecessary GHG emissions due to unit blowdowns to complete tests
in the not operating, depressurized mode. Annual three-mode testing will be difficult, if not
impossible, to coordinate with gas control pipeline operations. In addition, testing all modes in a
single site visit would be the likely approach. Increases in fuel (to bring on alternative
1550
compression) and unnecessary GHG vented emissions from compressor blowdown to address
reporting requirements is contrary to overall climate change objectives, and the Final Rule
should not introduce such scenarios. Operational changes for the sole purpose of making an
emission measurement for the remaining operating modes adds significantly to the burden of this
Proposed Rule, and poses significant scheduling issues for initial rule implementation.
Standby test crews or other opportunity approaches to test a compressor during periods
when the operating mode actually occurs are impractical due to logistics, service provider
availability, and/or extraordinary inefficiencies and implementation costs. Especially for initial
reporting years, multi-mode testing presents significant scheduling and resource issues (e.g., test
crew availability). The availability of qualified and experienced test personnel is not expected to
be sufficient to meet the demands during the initial years of the Final Rule and additional
demand from multi-mode testing would exacerbate this problem. The expected resource
limitations require a practical, phased-in approach or delay for the initial year of measurement
(as discussed in Comment XII).
Additional complexities are caused by vent accessibility issues. These include important safety
concerns that are discussed in Comment VI. In some cases, multiple compressor vents are
manifolded together, and measurement of the common vent line will not provide unit-specific
data by mode. However, manifolded vent line measurements can provide a reasonable, accurate
alternative for reporting annual vent emissions (see Comment VII).
In addition, forced-mode testing may not be indicative of actual emissions. For example, if the
common operating mode for a compressor is idle and depressurized, but three-mode annual tests
are required, the engine and compressor would likely be started from an idle mode to complete
the test. In this case, since the rod packing has not been energized for a considerable time, the
short-term emissions measured while test crews are on-site for a force-mode test may not be
indicative of typical emissions for that mode. Thus, the characterization test for the operating
mode could misrepresent emissions for that mode.
B. INGAA Recommends a Single Test Each Year in the As-Found Mode
In response to issues with three-mode testing, INGAA strongly recommends that annual tests be
conducted for a single as-found operating mode each year. This approach can be refined based
on EPA requirements. For example, the test could be completed in the prevalent operating mode
for that engine based on historical utilization records. This could be documented in the annual
report under 98.236 and the basis for the prevalent mode determination could be documented in
the GHG Monitoring Plan. INGAA recommends rule text revisions to 98.233(p)(4) to include
the following:
(4) Conduct one measurement for each compressor in each of the operational modes that occurs
during a reporting period its as-found operational mode. Document which of the three modes
was tested:
(i) Operating.;
(ii) Standby pressurized.; or
1551
(iii) Not operating, depressurized.

INGAA understands EPAs interest in mode-based testing is primarily due to vented emissions
from leakage through the unit isolation / block valve during the depressurized and standby,
pressurized compressor modes, respectively i.e., natural gas leakage past block valves when
the compressor is not operating. Emissions from leakage past pipeline unit isolation / block
valves are subsequently vented to atmosphere through the unit blowdown vent (i.e., unit
blowdown valve is open to atmosphere). As discussed in Comment V.D, the emission sources
and measurement or monitoring requirements related to these vented emissions needs to be
clarified, and considerable revisions may be required in the Final Rule to clearly address EPAs
intent. If INGAAs proposed alternative to annual three-mode testing is unacceptable to EPA,
INGAA requests additional discussion to ensure that a reasonable alternative to the Proposed
Rule requirements can be identified.
In employing the INGAA recommended approach, EPA will acquire data over time that provide
information on mode-based emissions without creating the significant problems described here.
In aggregate, mode-specific data gathered for this segment will facilitate an understanding of
mode-based emissions. As-found test data would continue to build upon existing modespecific operating data collected through the voluntary EPA Natural Gas STAR program, as well
as other available GHG emissions studies. These data should provide an adequate basis to inform
policy decisions.
EPA should also consider that operating time in standby pressurized mode is minimal. For most
operators in the gas transmission segment, the standby pressurized mode is a transient or
temporary mode, and company procedures typically limit time in this mode. This operating mode
exists after engine shutdown for a relatively short period, generally two to four hours, to
facilitate a short-duration maintenance activity or other near-term restart of the engine. Once the
allotted time period is reached, the pressure is released from the compressors, with the natural
gas vented or burned off as fuel. Regardless, cumulative time in this mode is short duration. This
action is driven by safety concerns, and the desire to release energy contained within pressurized
compressors out of the compressor building. As an example for annual implications, if a
company holds compressor pressure for four hours and releases pressure off the compressor 20
times throughout the year, this amounts to 80 annual hours in the standby pressurized mode or
less than one percent of the available annual time.
In addition, industry continues to implement programs to improve emission factors and reduce
the uncertainty of GHG emission estimates, including recent initiation of a project directed at
mode-based emissions. Testing is expected to begin in the summer of 2010 and data from the
program, in combination with INGAA member data and the existing knowledge noted above, are
expected to result in an improved understanding of mode-specific emissions. Thus, in the interim
until all three modes are tested, these data could be used to address data gaps, supplement data
collected from reporting, and provide interim access to mode-based data to inform this issue.
1552

Commenter:
Source Type:
Reciprocating compressor rod packing venting:
40 CFR 98.233(p)
1. Meter volumetric reading of gas emissions per unit time, under ambient conditions.
2. Total time the compressor associated with the venting is operational in the reporting year.
3. Mole percent of GHG i in the vent gas; use the appropriate gas compositions in paragraph
(u)(2) of this section.
Comment:
Compressor engines are generally either operational or not. A compressor would not standby
pressurized for any amount of time that would affect its emissions significantly as it would be
otherwise offset by the fact that the compressor is not operational. Therefore, requiring
measurement in each mode is burdensome and does not reflect a significant source of emissions.

Calculation methods for reciprocating compressor rod packing venting should be revised.
Section 98.232(d)(l) would require reporting from reciprocating compressor rod packing venting.
The calculation method for this source, prescribed at 98.233(p), has significant problems and
should be amended. To begin with, if compressor stations were improperly included as part of
the plant itself - a subject addressed above - then the prescribed calculation methods would be
extremely onerous. They would require the installation and maintenance of a huge number of
individual measuring units because such aggregation would result in so many individual source
points.
In addition, the prescribed calculation methods would force operators to conduct one
1553
measurement for each compressor in each of the operational modes that occurs during a
reporting period, i.e. operating, standby pressurized, and not operating, depressurized. This
would be a counter-productive requirement because not only would it force plants to shut down
in order to take such measurements, but it would also create greenhouse gas emissions that
would be created through the required depressurization (blowdown).
A preferable alternative to the methods currently set forth in 98.233(p) would be use of
the API Compendium for compressor measurements. The API Compendium would be
preferable because the compliance costs associated with the direct measurement requirements for
"large" reciprocating compressor rod packing vents are not justifiable given the relatively small
emissions that would be measured through use of such methods. Moreover, use of the API
Compendium would eliminate the current lack of clarity as to which reciprocating compressor
rod packing vents were "large" and which were not. See 75 Fed. Reg. 18620 (Table W-4).
subpart W; please see Section II.F of the preamble to todays final rule. EPA agrees to allow
compressor venting measurement in the as found mode with the exception of measuring
emissions from compressors in standby, depressurized at least once every three years. Please see
response to comment EPA-HQ-OAR-2009-0923-0055-16. In addition, the API compendium
factors do not account for emissions from unit isolation valves and therefore would not represent
all the emissions required under this rule. The commenter made an incorrect assumption that the
rule requires the installation of permanent flow meters on compressor vents. The rule allows for
temporary meters such as vane anemometers to measure compressor venting.

Organization:
(98.232 (c)(11)) Reciprocating compressor rod packing venting requires direct measurement
from open-ended vent lines when vent lines are present, and Optical Gas Imaging and direct
measurement when vent lines are not present. We propose emission factors and component
counts be used for emissions from compressor rod packings for onshore petroleum and natural
gas production, as this will significantly reduce the burden of leak detection and direct
measurement at wellsites and related production facilities.
Response: EPA considered comments received and determined that for onshore production for
both reciprocating and centrifugal compressors, that due to the small size of these compressors,
that population emission factors can be used, which would result in reduced burden on industry
while maintaining the necessary quality of data to inform policy, and todays final rule has been
revised accordingly. In todays final rule, onshore production reporters count the number of
centrifugal and reciprocating compressors and apply the appropriate emission factor. The other
sectors in the rule have compressors larger than those generally found in onshore production, as
well as larger emissions (please see the Technical Support Document EPA-HQ-OAR-2009
0923), therefore compressors in other sectors will not be allowed to use emission factors.
1554

Commenter:
Comment on 98.233(p)(2)(ii):
WBIH requests a clear definition of "all" vents. Does "all" specifically refer to rod packing,
blowdown valves, and blowdown vents?
Response: Reciprocating compressors vents include rod packing, unit isolation valves and
blowdown vents, and emissions manifolded to common vents.

Commenter:
Comment on 98.233(p)(4):
WBIH strongly requests measurement of emissions from rod packing, blowdown valves, and
blowdown vents be conducted in one mode, the mode existing at the time of the measurement. In
order to conduct measurements in the compressor de-pressurized mode, a compressor blowdown
would be required, resulting in additional emissions. Additionally, in the compressor de
pressurized mode, the compressor is not available for service creating an interruption of service
to customers
Please see response to comment EPA-HQ-OAR-2009-0923-0055-16.

The section on reciprocating compressor rod packing vents is one of the most burdensome
sections of the proposed rule due to the level of effort required to measure relatively small
emissions. EPA has greatly underestimated the cost and effort to conduct the direct measurement
prescribed for these sources. Aka has hundreds of compressors in operation, each one with
multiple compressor rod packings. Additional complexity and cost are presented in the many
cases when a single compressor is used for multiple services, for example inlet compression,
residue compression and refrigeration compression.
1555
Response: EPA has modified the rule to reduce the burden in gathering data from upstream
production compressor sources. Please see response to comment EPA-HQ-OAR-2009-0923
1063-1.

The direct method of measurement proposed for compressor rod packing vents is not valid in
many cases. A direct method will provide a snapshot which is unlikely to be valid as a
representative measure for an entire year. Alternative methods, such as documented and
controlled studies, would be more valid.
Response: EPA considered alternative methods of monitoring for reciprocating compressors
and determined that direct measurement minimized burden on industry while maintaining the
necessary quality of data to inform policy. Please see the Technical Support Document (EPA
HQ-OAR-2009-0923).

Section 98.233(p) Reciprocating compressor rod packing venting.
Onshore Production Source Category: Reciprocating compressor rod packing venting
The rule requires annual observation using an optical leak detection methodology followed by
measurement of the leak rate from each compressor seal, in each of three operating modes: (i)
operational; (ii) standby pressurized; and (iii) not operating, depressurized.
This approach poses numerous concerns:
- Each compressor will need to be monitored when it is in the various modes during the year in
order to arrange for monitoring at the specified conditions. This presents difficulties in
scheduling a contractor experienced in the use of the optical leak detection equipment, or
requires the operator to have enough of these devices and trained users on hand and in proximity
to the compressors when needed (the infrared cameras cost up to $100,000 each to purchase, or
about $2,500 per week to rent).
- Alternatively, an operator will need to purposely conduct the compressor shut down, isolation,
depressuring, and restart specifically for fulfilling the reporting requirements. Depressuring
every compressor each year, in itself, will result in unnecessary greenhouse gas emissions and
1556
presents a safety concern for the release of a combustible gas.

- There is an economic concern in shutting down compressors that are necessary for production
operations. The rule provides no option for the use of best available monitoring methods if a
compressor shut down cannot be scheduled for these measurements.
- Operators will be required to track cumulative time per year for each compressor in the three
modes of operation, a significant burden for dispersed equipment. API believes that EPA has
significantly underestimated the number of compressors associated with onshore oil and gas
production operations by neglecting to account for well-head compressors, resulting in some
33,000 compressors impacted by the rule, and therefore the costs associated with this emission
source are significantly underestimated.
- Compressor vent-line access may pose an unnecessary safety risk. In many cases, the design of
existing facilities prevents safe access to compressor vents. In addition, accessing vents could
place technicians at risk should an emergency shutdown (i.e., blowdown) occur while conducting
a vent measurement. While alternative access may be feasible at some facilities, the rule
implementation schedule prevents design and installation of vent access points at all facilities
within the prescribed timeframe for initial reporting.
As an alternative estimation method for compressor rod packing venting, API proposes that the
emission rate determination be conducted as follows for a representative sampling of
compressors in their as-found mode rather than the 3 operating modes specified above. Over
time, data will be gathered for groupings of similar compressors in each of the three specified
modes. In aggregate, data from all operators should facilitate an understanding of mode-based
emissions.
This representative sample would be comprised of a percentage of the reciprocating compressors
for a Sub-basin entity taking into account the safety risk described above. The representative
sample would be specifically defined in the GHG Monitoring Plan for a Basin entity and it
would be in the range of 1 to 10 percent of the reciprocating compressors for a Sub-basin entity
during a year. This representative sample of measured emissions from reciprocating compressor
rod packing would then be averaged and applied as the emission factor for all other compressors
in the Sub-basin entity.
In general, the representative sample should be evenly distributed based on age of the
reciprocating compressors, where the distribution will be determined by the age of a Sub-basin
entitys actual compressor fleet.
Once a Sub-basin entity has performed direct measurement on at least 20 percent of the total
number of reciprocating compressors in their as-found mode over a period not to exceed 3
years, no further direct measurement will be done and the average of the collected emission data
would be used as the emission factor for the Sub-basin entitys reciprocating compressors from
that point forward.
1557
Response: EPA will allow the use of population emission factors for compressors in onshore
production. Please see response to comment EPA-HQ-OAR-2009-0923-1063-1.

Section 98.233(p)(2)(ii) Reciprocating compressor rod packing venting. An orifice meter is
inappropriate for use in measurements for this scenario due to low accuracy at low flows and
pressure, furthermore, the added back-pressure is a safety and reliability concern.
Response: EPA does not require an orifice meter for reciprocating compressor venting
measurement, and it would be the responsibility of the reporter to determine the safe operations
of their facilities.

Section 98.233(p) Reciprocating compressor rod packing venting. The variable Mi in Equation
W-17 should be a volume percent of GHG to make the units work out.
Response: EPA disagrees with this comment to change the variable Mi in this equation to
volume percent. Please see the response to comment EPA-HQ-OAR-2009-0923-1151-81.

Section 98.232(c)(11) and Section 98.233(p): For reciprocating compressors, the rule requires
annual observation using an optical leak detection methodology followed by measurement of the
emissions rate from each compressor seal (equivalent to the number of compression cylinders)
using a variety of methods. Based on a range of one well-head compressor per 12 gas wells to
one well-head compressor per 20 gas wells, a count of compressors associated with oil
production from EPAs national inventory, and the assumption that 81% of equipment are
covered by the rule, there are between 20,000 and 33,000 well-head compressors subject to the
rule. With a range of 2 to 4 cylinders per compressor and the fact that almost all well-head
compressors do not have common seal collection piping in place, this yields between 40,000 and
132,000 seals which will require measurement. Additionally, such measurement is required in 3
separate operational modes (operating, stand-by pressurized, and stand-by depressurized)
1558
yielding an estimated range of 122,000 to 396,000 separate required measurements. Assuming

that it takes a technician 4 hours to inspect the compressor with an optical device, measure
emissions from each seal (most likely using a bagging methodology), shut-down the compressor
in a pressurized mode and repeat the measurements, then depressure the machine (which will
yield significant emissions) and repeat the measurements, an average of 107,000 hours will be
required. At an estimated cost of $150 per hour for a specialty contractor with an optical device
and leak measurement equipment and $100 per hour for an operator to accompany the specialty
contractor and handle the compressor shut down, isolation, depressuring, and restart, this yields a
projected cost ranging from $20 MM to $33 MM. The costs to install seal leak collection
systems will far exceed the costs projected above and are not expected to be widely adopted in
the short term.
Response: EPA has revised todays final rule and does not require direct measurement of
reciprocating compressor venting in onshore production. Please see response to comment EPA
HQ-OAR-2009-0923-1063-1.

Organization: Southern Ute Growth Fund
Commenter: Lynn Woomer
Section, 98.233(p) Reciprocating Compressor Rod Packing Vents, proposes direct
measurements. In particular, the proposed rule will require measurements of a single compressor
in three distinct modes - (1) while operating, (2) while on standby, and (3) when depressurized.
This triple measurement scenario on a single unit is neither reasonable, nor justifiable to acquire
at most, minimal emissions data. The SUGF suggests EPA reconsider usage of the three
measurement scenarios; in particular, eliminate measurement of a depressurized unit. This will
lessen the resources needed to acquire data, yet remain consistent with other CAA program (Le.
a unit is not required to start up or shut down solely for the purpose of conducting an emissions
test).
Response: EPA agrees to allow compressor venting measurement in the as found mode, see

Section 98.233(p) Reciprocating compressor rod packing venting.
The rule requires annual observation using an optical leak detection methodology followed by
measurement of the leak rate from each compressor seal, in each of three operating modes: (i)
1559
operational; (ii) standby pressurized; and (iii) not operating, depressurized.

This approach poses numerous concerns:
- Either each compressor will need to be monitored for when it is in the various modes during the
year in order to arrange for monitoring at the specified conditions. This presents difficulties in
scheduling a contractor experienced in the use of the optical leak detection equipment, or
requires the operator to have enough of these devices and trained users on hand and in proximity
to the compressors when needed (the infrared cameras cost between $80,000 and $100,000 each
to purchase, or about $2,500 per week to rent).
- Alternatively, an operator will need to purposely conduct the compressor shut down, isolation,
depressuring, and restart specifically for fulfilling the reporting requirements. Depressuring
every compressor each year, in itself, will result in unnecessary greenhouse gas emissions and
presents a safety concern for the release of a combustible gas.
- There is an economic concern in shutting down compressors that are necessary for production
operations. The rule provides no option for the use of best available monitoring methods if a
compressor shut down cannot be scheduled for these measurements.
- Operators will also be required to track cumulative time per year for each of the three modes of
operation, which is a significant burden for dispersed equipment and not very feasible. BP
believes that EPA has significantly underestimated the number of compressors associated with
onshore oil and gas production operations by neglecting to account for well-head compressors
which the rule appears to cover. As an alternative estimation method for compressor rod packing
venting, BP proposes that emission rate determination be conducted for a representative
sampling of compressors in their primary operating mode in the sub-basin entity. Over time, data
will be gathered for groupings of similar compressors in each of the three specified modes. In
aggregate, data from all operators should facilitate an understanding of mode-based emissions.
BP requests that reporting emissions during normal operation only be required. While the
greatest emission rate occurs during pressurized standby operation, the amount of time spent in
this operating mode is insignificant, amounting to a small number of hours per year. Our
reciprocating compressors are in operation most of the year and if not operating these machines
are quickly brought to a depressurized shut down state. Therefore, the only time the machines are
in pressurized standby is when the machine is transitioning between normal operation and shut
down. As these machines are rarely shut down and this transition period (pressurized standby is
brief) accounting for this specific operation mode should not be required. In addition, emissions
during depressurized shut down will be negligible as the process gas is not pressurized and does
not have proper force/motive to leak process gas through the rod packing; therefore, measuring
and reporting emissions during this mode of operation should be excluded from the rule.
Response: EPA will allow the use of population emission factors for compressors in onshore
production. Please see response to comment EPA-HQ-OAR-2009-0923-1063-1.
1560

Compressor Seal Systems (Rod Packings on Reciprocating Engines, Dry/Wet Seals on Turbines)
Compressor seal systems are designed to keep high pressure gas from escaping through a rod
into the distance piece (for reciprocating engines). Without going into great detail, these systems
are designed to leak a little and over time will have to be replaced because of constant wear on
the compressor rod. Because of the nature of these systems, venting gas that does escape through
the packing is channeled either directly outside of the building through a separate manifolded rod
packing line or directed into the distance piece where typically a larger diameter pipe is
manifolded with other compressors to vent this gas out through the roof line in a safe manner.
In order to save valuable time, we recommend that direct measurement with a high volume
sampler be performed at each of these vents in order to determine an accurate leak rate. To
perform leak screening with a gas imaging camera on these systems is simply a waste of time
and adds no value to the process. The same is true with conventional leak detectors as they will
typically peg out full scale with the presence of natural gas at each vent.
Safety is still paramount to performing these measurements as it may require the use of Fall
Protection and a firmly grounded and well positioned man-lift or ladder. Training in
measurement, no matter what method is attempted, should also be given consideration as not all
systems are designed the same. With regard to packing systems for reciprocating engines alone,
there should be clearly established protocols and standards in place to insure a level of
consistency, accuracy and safety is achieved. Again, the safest, most accurate and most reliable
method for this leak measurement in our opinion is with the use of a High Volume Sampler.
13.2.8
OFFSHORE EMISSIONS CALCULATIONS (GOADS SYSTEM)

Organization: Innovative Environmental Solutions
Commenter: Thomas McGrath
98.232(b) refers to all "stationary fugitive" and "stationary vented" sources as identified in an
MMS GOADS study. Please provide specific lists of these sources with descriptions/definitions
for Subpart W.
Response: EPA has revised todays final rule to require reporting of platform equipment leaks,
vented, and flare emissions as identified in the data collection and emissions estimation study
1561
conducted by BOEMRE in compliance with 30 CFR 250.302 through 304. Offshore platform
operators subject to GOADS reporting, shall include those applicable sources which are listed in
the GOADS activity data request software, and for which BOEMRE calculated and published
GHG emissions. Reporters in state and non-Gulf of Mexico waters will report applicable
sources under subpart W using GOADS methodologies. EPA has not provided a list of sources
covered by the BOEMRE GOADS study or descriptions and definitions in todays final rule
subpart W because the information is incorporated by reference and available publicly through
the BOEMRE Gulfwide Emissions Inventory Study.

Organization: Innovative Environmental Solutions
Commenter: Thomas McGrath
The MMS GOADS study collects monthly activity data and calculates emissions for the
following sources:
Amine units Boilers/heaters/burners Diesel and gasoline engines Drilling rigs Combustion flares
Fugitive sources Glycol dehydrators Losses from flashing Mud degassing Natural gas engines
Natural gas turbines Pneumatic pumps Pressure/level controllers Storage tanks Cold vents
Emissions are calculated at the platform and unit level (e.g., boiler, engine, etc). For more
information on the GOADS study, please see the U.S. Department of the Interior, Minerals
Management Service Website:
http://www.gomr.mms.gov/homepg/regulate/environ/airquality/goads -2008.html.
Response: EPA reviewed this comment and has retained GOADS data collection methodologies
for offshore platforms.
The proposed Subpart W is not very clear about the requirements for the offshore petroleum and
gas production facilities. The rule should specify whether the facilities for which emissions are
calculated under the MMS GOADS (Gulfwide Offshore Activities Data System) program report
emissions only for the years for which MMS collects the GOADS activities data from the
operators, or whether the offshore facilities perform their own calculations using the GOADS
activity data and GOADS emission factors and report their emissions annually to EPA according
to 40 CFR 98.2(a) (I). The final rule should clearly state that GOADS is just the software
program for collecting activities data for the facility; it does not calculate emissions. The
activities data are then fed into a data base system, which contains the emission factors for
calculating emissions. The methodologies for calculating emissions are presented in the report
1562
published by MMS that presents a summarization of the emissions.

The draft rule 98.233 is organized in such a way that it is difficult to discern which
requirements are for offshore facilities only, which ones are for onshore facilities only, and
which apply to both classes. This needs correction in the final rule.
Response: Under Subpart W, offshore reporters must report every calendar year starting with
2011 using the latest GOADS emissions data (for GOADS reporters); and calculated emissions
based on the latest published methodologies (for non-GOADS reporters). However, reporters are
required to adjust emissions each year (between GOADS cycles) based on the operating time for
the facility relative to operating time in the previous reporting period. The GOADS software,
and the Data Base Management System (DBMS), are used by BOEMRE to calculate emissions
from GOADS activity data.
Page 18612, 1st column, 1st paragraph: It should be noted that the GOADS software is an
activity database system to provide the necessary data to calculate emissions. The software itself
does not calculate emissions. The operator of an offshore facility enters information into the
GOADS software, which is delivered to MMS. After the information is deemed complete, it is
sent to a contractor, where the data are compiled into a data base management system (DBMS),
which calculates emissions based on the activity data and emission factors. However, all the
formulas for calculation of the emissions are given in the GOADS report. Please clarify in the
rule that in order to calculate emissions, the operators would also need to use the GOADS study
emission factors. Information about GOADS and the emissions inventories may be found at:
http://www.gomr. mms.gov/homepg/regulate/environ/airquality/goads.html .
In the 6th line from the top of the page, it may be more appropriate to state "(e.g., California and
Alaska)" rather than "(i.e., California and Alaska)" since EPA would presumably recommend the
GOADS reporting method for any possible future facilities in other areas, such as in the Atlantic
Ocean.
Response: EPA agrees with these comments. For more information see response to comment
EPA-HQ-OAR-2009-0923-0847-1.
Page 18643, 233(s) (l). MMS GOADS Reporters: It should be clarified again that GOADS is
the software for reporting activity data; it does not calculate emissions. Emissions are calculated
using the methodologies in the GOADS emission inventory report. There are two "(ii)" under
1563
233(s)( I). The second "(ii)" should be changed to "(iii)" and "(iii)" to "(iv)". For each reporting
year that does not overlap with the GOADS reporting year, emissions must be adjusted based on
the operating time for each platform. Please clarify how the emissions should be adjusted.
Operating time is not the only variable as there are other factors that can change emissions.
Page 18643, 3rd column. last item under 233(s)(l): The subparagraph states "If MMS
discontinues or delays their GOADS survey by more than 4 years, then Platform operators shall
collect monthly activity data every 4 years from platform sources . . . ." The GOADS survey is
conducted every 3 years. It is not obvious why a time interval of 4 years is chosen. This is
inconsistent with 98.2(a) (I) which requires annual GHG reporting. The same comment applies
to 233(s)(2), Non-MMS GOADS Reporters.
Response: EPA determined to use operating time of the facility as the only variable by which to
adjust emissions between GOADS cycles, to reduce burden. Subpart W requires updating the
reported emissions each time a new GOADS emissions data and methodologies are published,
unless BOEMRE discontinues or delays their GOADS survey by more than 4 years. In that case,
offshore reporters are required to calculate emissions according to the latest published GOADS
methodologies and emission factors every 4 years, and report those under subpart W for
subsequent years, i.e. the reporters will calculate detailed emissions every 4 years, and for the
intermediate years the reporters would adjust their platform emissions using an operating factor.
EPA determined to use 4 years, as BOEMRE has generally conducted the GOADS survey every
3 years, and therefore todays final rule would not in general supersede the GOADS cycle. For
more information see response to comment EPA-HQ-OAR-2009-0923-0847-1.
98.235: A complete record of all estimated and/or measured parameters used in the GHG
emissions calculations is required. If data are lost or an error occurs during annual emissions
estimation or measurements, you must repeat the estimation or measurement activity for those
sources as soon as possible, including in the subsequent reporting year if missing data are not
discovered until after December 31 of the reporting year, until valid data for reporting is
obtained. Data developed and/or collected in a subsequent reporting year to substitute for
missing data cannot be used for that subsequent years emissions estimation. Where missing data
procedures are used for the previous year, at least 30 days must separate emissions estimation or
measurements for the previous year and emissions estimation or measurements for the current
year of data collection.
OOC Comment: The rule should state that section 98.235 does not apply to offshore platforms.
The MMS GOADS process allows for the use of surrogate data in some circumstances when
data is not available. Thus, to prevent confusion, the GOADS procedures should be used to
handle missing data.
On a general basis, what do you do if missing data comes up in the last 2-3 weeks of December
1564
of the previous year, and again in early January of the current year? EPA should remove or
reword this sentence.
Response: EPA agrees that those platforms required to report to BOEMRE in a GOADS study
should follow those procedures for missing data. However, for platforms that do not report to a
BOEMRE GOADS study, it is incumbent on the platform owner/operator to follow the subpart
W procedures for estimating missing data outlined in 98.235. Regarding the general comment
that certain data is available only in the last weeks of one year and the first weeks of the
following year, EPA disagrees that the missing data procedures should be removed. It is
incumbent on the owner/operator to make every effort to collect and retain required data. Data
available in the last weeks of a reporting year, and then again only in the first weeks of the
following year must be collected and retained for each period and each year when the
opportunity to collect it arises. Section VI of the preamble to The Final Mandatory GHG
Reporting Rule (Final MRR), (40 CFR part 98), states that merely filling in missing data does
not excuse a failure to perform the monitoring or testing. Filling in data gaps that are missing
does not relieve an operator from liability for failure to continuously monitor and test as
required, as discussed in the response to comment EPA-HQ-OAR-2008-0508-0493.1, excerpt
40.
98.232(b): For offshore petroleum and natural gas production, report emissions from all
stationary fugitive and stationary vented sources as identified in the Minerals Management
Service (MMS) Gulfwide Offshore Activity Data System (GOADS) study (2005 Gulfwide
Emission Inventory Study MMS 2007-067). Also, additional information on utilizing GOADS in
98.233(s) will be commented on and/or questioned.
OOC Comment: MMS Study 2007-067 does not use the terms stationary vented or stationary
fugitives. This leads to confusion about what emission sources from GOADS are included.
Currently, we only provide an equipment inventory, operating times and fuel usage to the MMS,
and they run the emissions calculations. We are currently awaiting results from the most recent
GOADS study, and the wait could routinely be over the 3 months time we are given to report
emissions. For non-MMS GOADS reporters, we will need to obtain the latest MMS GOADS
emission factors and methods or find another method for calculating venting and fugitive
emissions, as these are not part of the MMS GOADS inventory we provide, nor the final
emissions report we receive from the MMS. Is there an agreement between the MMS and EPA
on how to manage this interface and any other coordination issues? Is this the latest GOADS
study available? If there are other programs in place for GHG emissions calculating and
reporting in non-MMS GOADS offshore areas (such as state or voluntary programs), can we
have the option to use these instead?
1565
Response: Regarding the terms stationary vented or stationary fugitives, please see
response to comment EMAIL-0010-1 (comment also located in rulemaking memo Early
Comment Submissions in docket EPA-HQ-OAR-2009-0923). Regarding reporters in state and
non-Gulf of Mexico waters, please see response to comment EMAIL-0010-1 (comment also
located in rulemaking memo Early Comment Submissions in docket EPA-HQ-OAR-2009
0923). Should the 2011 Gulfwide Emissions Inventory Study not be published in time for
reporting year 2011, then for reporting year 2011, offshore reporters subject to GOADS must use
the latest (2008) GOADS emissions data. Reporters in state and non-Gulf of Mexico waters (not
required to report under GOADS) will report under subpart W for 2011 using the GOADS
methodologies published with the 2008 GOADS emissions data. In subsequent years, reporters
in state and non-Gulf of Mexico waters will report using the most recent GOADS methodologies
published, which historically have been published along with the GOADS emissions data. EPA
is not allowing in todays final rule the use of any other methodologies for offshore platform
emissions calculation and reporting.
98.232(k): You must report under Subpart C of this part (General Stationary Fuel Combustion
Sources) the emissions of CO2, CH4, and N2O from each stationary fuel combustion unit by
following the requirements of Subpart C. Also, 98.2(a)(3)(ii) states that the aggregate maximum
rated heat input capacity of the stationary fuel combustion units at the facility is 30 mmBtu/hr or
greater.
OOC Comment:
Since Section 98.232(b) already requires reporting all stationary fugitive and stationary
vented sources from GOADS, GOADS should be used for reporting combustion emissions, as
this information is already collected for GOADS and would eliminate duplicate reporting.
By using GOADS for reporting of combustion emissions, EPA should state that offshore
platforms are exempt from reporting under subpart C.
Response: EPA disagrees that GOADS should be used for reporting all combustion emissions.
For offshore production, only GOADS flare combustion is reported under subpart W. All other
applicable offshore combustion sources must follow the appropriate data collection requirements
of subpart C, beginning in 2010. Please see response to comment EMAIL-0010-1 (comment
2009-0923).
1566

Section 98.232(j) and 98.232(k):
(j) You must report the CO2, CH4, and N2O emissions from each flare.
OOC Comment: It is not clear whether this section applies to offshore platforms. If flare sources
are to be included, the language should be modified to prevent confusion. Since Section
98.232(b) already requires reporting all stationary fugitive and stationary vented sources
from GOADS, it would be less confusing to use GOADS for flare emissions, as this information
is already collected for GOADS and would eliminate duplicate reporting.
Response: EPA agrees that offshore production must report to subpart W their flare emissions
using GOADS. Please see response to comment EMAIL-0010-1 (comment also located in
Use of OGI Should be Required at Additional Facilities, Especially Offshore Petroleum and
Natural Gas Production Facilities
FLIR Systems believes that the MMS GOADS methodology recommended by the EPA for
emissions on offshore oil platforms will inaccurately report emissions from these locations.
Turbulent wind conditions make it difficult for operators to properly locate and consistently
measure leaks on an oil rig with TVA technology. The video linked above, collected from an
oilrig in April of this year, makes this quite clear. In addition, the MMS GOADS
recommendation will create safety hazards that will compromise worker safety. The video linked
below, taken at the same time, shows a large gas plume wafting through the grating on the same
platform.
http://www.flir.com/thermography/americas/us/OGI/offshoreplatforms/
The risks of explosions do exist on oil rigs can be reduced with the proper deployment of OGI.
At minimum, annual scans of oil platforms to identify leaks should be required.
Response: The commenter made assumptions which are not consistent with the proposed rules
intent. EPA never intended to specify methodologies for BOEMRE GOADS studies and todays
final rule requires reporting GHG emissions based on BOEMRE Air Emissions Inventory Study
methods as defined by BOEMRE in their GOADS data system and DBMS emissions calculation
and reporting software.

1567

We do not support the use of the Minerals Management Service (MMS) Gulfwide Offshore
Activities Data System (GOADS) air quality data collection system for tracking and reporting
GHG emissions from offshore oil and gas sources. Nor do we support expanding the [GOADS]
system to platforms in other Federal regions (e.g., in California and Alaska) and all State
waters 231 .
We consider this to be an inferior standard of measurement for offshore sources.
Response: EPAs decision to adopt BOEMRE GOADS methodology is cost-effective in not
requiring duplication of effort, and extension of BOEMRE GOADS methodologies to offshore
platforms in state and non-Gulf of Mexico waters provides for more complete coverage and
consistent reporting of all offshore operations above the reporting threshold and therefore EPA is
retaining GOADS reporting for offshore reporters.
Organization: Alaska Oil and Gas Association
Commenter: Marilyn Crockett
Subsection 98.232(b) contains the following requirement: For offshore petroleum and natural
gas production, report emissions from all stationary fugitive and stationary vented sources
as identified in the Minerals Management Service (MMS) Gulfwide Offshore Activity Data
System (GOADS) study (2005 Gulfwide Emission Inventory Study MMS 2007 067).
Companies operating offshore platforms in State waters in Alaskas Cook Inlet have no
experience or systems in place to monitor and report emissions under the GOADS system and
the proposed rule does not clearly identify sources subject to reporting. Therefore, we
recommend that EPA allow the use of best available monitoring methods for the first year of data
collection for operations occurring offshore in Cook Inlet.
Response: EPA agrees that use of best available monitoring methods may be necessary for
some reporters for the 2011 reporting period so long as the reporters meet the best available
monitoring method criteria as described in the rule. Reporters may apply for the use of best
available monitoring methods by February 28, 2011 and cite the specific sources and monitoring
requirements that cannot be met and justify to EPAs satisfaction why full compliance cannot be
met. Please see Section II.F.4 of the preamble to todays final rule, and response to EMAIL
0010-1 (comment also located in rulemaking memo Early Comment Submissions in docket
EPA-HQ-OAR-2009-0923).
231
See 75 FR 18611, April 12, 2010.

1568

Offshore Oil and Gas Production Source Category Specific Comments
BP supports EPAs reliance on the Gulfwide Offshore Activity Data System (GOADS) study for
emission reporting from the Offshore Petroleum and Natural Gas Production source category.
However, as stated in the comments from the Offshore Operators Committee (OOC), there are
several areas where additional clarity is needed. BP supports the comments submitted by the
OOC.
Response: For further information, please see response to comments EPA-HQ-OAR-2009-0923
1027-7, EPA-HQ-OAR-2009-0923-1027-13, EPA-HQ-OAR-2009-0923-1027-8, and EPA-HQ
OAR-2009-0923-1027-9.
13.2.8.1
FEDERAL WATERS
There are no comments under this category.
13.2.8.2
STATE WATERS

Section 98.233(s) Offshore petroleum and natural gas production facilities in both state and
federal water. API supports the use of GOADS data for offshore petroleum and natural gas
production facilities. Other programs besides GOADS currently exist under which offshore
facilities report emissions, such as state programs. For example, Alabama requires offshore
facilities to report emissions. API recommends for non-MMS GOADS reporters, an alternative
be added that allows offshore operators to report the information that is submitted to state
agencies.
Response: EPA disagrees that facilities in state waters should be allowed to use inventories
developed for the state for reporting to this rule. EPA considered allowing alternatives such as
state inventories, but decided it was important to receive data collected using consistent methods
for this source. Allowing reporters to submit state inventories may result in a loss of data for
several activities and additionally will allow for variations in emissions estimates quality because
of the different methodologies. Further, having a uniform requirement across the entire industry
diffuses potential confusion over the applicability of subpart W and which emission sources are
included in the 25,000 tonne CO2e threshold.
1569
13.2.9
CALCULATION OF VOLUMETRIC AND MASS EMISSIONS

98.233 (t) Equations do not currently use (compressibility factor). Should this be incorporated
into the equations
Response: EPA disagrees that use of the compressibility factor is necessary to determine
emissions of CH4 and CO2 from the petroleum and natural gas industry. At most operational
conditions expected, these two species will acceptably exhibit near-ideal gas behavior, such that
any small error in the calculation will not be expected to substantially impact the results of the
emissions calculations. .

Section 98.233(u)(2)(iv) GHG Volumetric emissions for underground natural gas storage
facilities
The proposed rule requires that GHG volumetric emissions be calculated at standard conditions
using the natural gas volumetric emissions at standard conditions (Es,n) and mole percent of
GHG in the natural gas (Mi). For underground natural gas storage facilities, Mi shall be the
annual average mole percent in the natural gas stored underground. However, natural gas
stored typically includes base gas and newly injected gas. Gas composition may not be readily
available for base gas in older storage facilities and the composition of the injected gas will only
be known for each injection cycle. Fugitive emissions for storage stations will occur from
sources located between the suction header and storage wells, El Paso recommends the gas that
passes through the facility be allowed as the basis of the GHG mole percent for underground
natural gas storage facilities.
Response: EPA agrees that GHG volumetric emissions for underground natural gas storage
facilities may be based on the mole percent of GHG in the natural gas that passes through the
facility. EPA understands that sampling of the natural gas for composition analysis is done at
these facilities, and not from the underground well. Additionally, EPA understands that most
emissions from this industry segment will occur from the above-ground compression and
processing equipment, not the storage wellhead. Thus, todays final rule clarifies that to estimate
GHG volumetric emissions for underground natural gas storage facilities, reporters will use the
GHG mole percent in natural gas that passes through the facility.
1570

Proposed Rule Section 98.233(u)(2)(iv) GHG Volumetric emissions for underground natural gas
storage facilities.El Paso requests GHG volumetric emissions for underground natural gas
storage facilities be based on mole percent of GHG in the natural gas that passes through facility.
Response: EPA agrees that reporters may use the mole percent of GHG in the natural gas that
passes through underground natural gas storage facilities to represent the volumetric emissions
from natural gas storage facilities. Please see the response to comment EPA-HQ-OAR-2009
0923-1011-49.

Improve Consistency
API requests that EPA conform the definitions and terminology used throughout the proposed
rule to make it more consistent with that used by industry. An example of the use of terms that
are not those commonly used by industry or equipment vendors, is the rules use of high bleed
devices rather than high bleed controllers, which is the term recognizable to equipment
vendors. APIs specific definition recommendations are provided in Section III.1 of this
document.
However, even more fundamental than the terminology issues, is the proposed rules confusing
mix of standard conditions for expressing gas volumes and mass. The MRR uses, in different
parts of the existing rule (Subparts C, Y, NN, PP) as well as the proposed Subpart W, industry
conditions, ambient conditions, actual conditions and STP. API strongly requests that the rule
should consistently apply industry standard conditions (60 degrees F and 14.7 psia) since these
are the units used in all the referenced industry standards and are the common units used for
calibrating industry devices for custody transfer.
In addition, applying the emission equations in the proposed Subpart W produces a mix of
emission results. For example, Equation W-1 results in natural gas emissions (scf/yr) from high
bleed pneumatic controllers, while Equation W-2 results in annual metric tonnes of CO2
equivalent emissions from low bleed pneumatic controllers. Although, Section 98.233(u)
Equation W-22 converts natural gas volumetric emissions to GHG volumetric emissions and
Section 98.233(v) Equation W-23 converts GHG volumetric emissions to GHG mass emissions
as CO2e, the mix of units resulting for individual source types impedes comparison among
1571
sources. For consistency with existing MRR reporting requirements and for comparison among
emission source types, all emission equations should result in metric tonnes of GHG emissions
by GHG type (e.g., CO2, CH4, and N2O) and then apply Equation A-1 to convert to CO2e
emissions.
Consistent use of industry standard conditions for all calculations with conversion of volumes to
metric tonnes for each quantified source will ensure that annual GHG emissions are properly
quantified and reported.
Response: EPA disagrees that terminology used throughout the rule is not common. The
terminology is consistent with partner companies in the Natural Gas STAR Program and other
EPA experience and regulations. Subpart W uses the standard conditions of 68 degrees
Fahrenheit and 14.7 pounds per square inch absolute as defined in subpart A, and provides
calculation methods to convert the reporters temperature and pressure data to standard
conditions. However, EPA disagrees that each equation under subpart W must result in
emissions of the same units. The methodologies prescribe that the results of each equation must
be converted into standard volumetric emissions and mass emissions using provided conversion
equations, and this would ensure reporting on a common basis across sectors in the MRR.
Section 98.233(t)(1) Volumetric emissions. The value of 459.67 should be utilized in place of
460 for equation W-20. However, as stated in earlier comments, API recommends quantifying
emissions in subpart W in terms of tonnes to avoid potential errors in converting between
volume and mass for gas phase emissions, particularly when the rule applies different sets of
standard conditions for gas streams.
Response: EPA disagrees that using the hundredths decimal place in the conversion from
degrees Fahrenheit to degrees Rankine is necessary for converting volumetric emissions to
standard conditions since this would introduce more than necessary precision and burden
(reporting accurately to the hundredths decimal place) and also the resulting difference would be
expected to be insignificant. Additionally, EPA disagrees that each equation under subpart W
must result in emissions of the same units of tones. Please see the response to EPA-HQ-OAR
2009-0923-1151-8.

Section 98.233(u)(i) GHG volumetric emissions. In determining the GHG mole percent in
produced natural gas for onshore petroleum and natural gas production facilities, the rule states:
1572
If you have a continuous gas composition analyzer for produced natural gas, you must use these
values in calculating emissions. If you do not have a continuous gas composition analyzer, then
quarterly samples must be taken according to methods set forth in Section 98.234(b). This
appears to require quarterly sampling and analysis for each well. If measurement is required at
each well - there are approximately 667,000 wells impacted by the rule 4 samples/year $200
each (which is on the low side). This equates to $534 MM per year just for the sampling, or
about 20 times the total first year cost in the EIA.
In order to reduce the cost burden, API recommends that the final rule should explicitly state that
the composition data/GHG mole percent can be obtained from representative annual samples of
produced natural gas within the Sub-basin. Consistent with APIs proposed Sub-basin entity
approach, the representativeness of the samples can be based on grouping of fields (in EIAs
field list) within a Basin entity based on common production characteristics. These
characteristics would be the same (or groups of similar) producing reservoirs, similar well depths
and well-bore configurations, similar pressure and temperature ranges, similar fluid compositions
and GORs, similar production arrangements and surface equipment, and similar operational
characteristics and practices. The Basin entity should discuss the produced natural gas sampling
strategy in the monitoring plan.
Response: EPA agrees that reporters may use representative samples of field gas composition
and that quarterly sampling of produced gas composition is not necessary. Todays final rule
clarifies that annual sampling of produced natural gas for composition analysis is allowed or that
reporters may use the most recent representative sample analysis of the field that is already
available for the purposes of subpart W. Please see Section II.E of the preamble to todays final
rule.
Section 98.233(u)(i) GHG volumetric emissions.
In determining the GHG mole percent in produced natural gas for onshore petroleum and natural
gas production facilities, the rule states: If you have a continuous gas composition analyzer for
produced natural gas, you must use these values in calculating emissions. If you do not have a
continuous gas composition analyzer, then quarterly samples must be taken according to
methods set forth in Section 98.234(b). It is not clear if sampling must be conducted quarterly at
each well or if quarterly samples need to be taken at the basin level which implies just 4 samples
per operator per basin annually. If sampling and analysis is required at each well the sampling
burden is unreasonable and not necessary.
In order to reduce the cost burden, BP recommends that the final rule should explicitly state that
the composition data/GHG mole percent can be obtained from representative annual samples of
produced natural gas within the sub-basin. Consistent with APIs proposed sub-basin approach
the representativeness of the samples can be based on grouping of fields (in EIAs field list)
1573
within a basin based on common production characteristics. As the source categories included in
Subpart W all produce/process fluids from fluid reservoirs, fluid properties will be consistent
throughout the year for each field and there will be insignificant variance amongst quarterly
samples. This contrasts refinery operation which may receive different crude feed streams and
where there may be a variance in composition analysis between quarters. Reducing the sampling
requirement from quarterly and individual locations to annual and representative samples will
provide equal quality information to meet EPAs intent while reducing the burden to gather this
information. The reporting entity should discuss the produced natural gas sampling strategy in
the monitoring plan.
Response: EPA agrees that requiring quarterly sampling of produced gas from each well in a
basin for composition analysis is not necessary; please see the response to EPA-HQ-OAR-2009
0923-1151-50.
Section 98.233(u) GHG volumetric emissions. The variable Mi in Equation W-22 should be a
volume percent of GHG to make the units work out.
Response: In todays final rule, EPA more correctly expresses this term in mole fraction, which
is close enough to volume fraction so as to be considered synonymous assuming near-ideal gas
behavior; please see the response to EPA-HQ-OAR-2009-0923-1151-81.

GHG volumetric emissions
Section 98.233(u): GHG volumetric emissions generally would require the use of continuous gas
analyzer or quarterly gas analyses for facilities operated by IPAMS member companies. There
is no justification for such frequent gas analyses. Therefore, IPAMS requests that a one-time
analysis at the field level be used for this calculation.
Response: EPA disagrees that a one-time composition analysis at the field level will be
sufficient to characterize the composition of produced natural gas for that field indefinitely.
While EPA agrees that quarterly sampling is not necessary, studies have shown that over years
composition of aging wells and fields change. Thus, todays final rule stipulates that the
composition of produced natural gas for each field can be based on the most recent available
sample analyses; Please see the response to EPA-HQ-OAR-2009-0923-1151-50. For more
1574
information, please see Section II.E of the preamble to todays final rule.
13.2.10
EOR INJECTION PUMP BLOWDOWN

EOR Injection Pump Blowdown
Source Description: EOR operations where CO2 is injected into a formation to increase
production use pumps to inject CO2. These pumps are periodically blowndown to perform
maintenance and repair activities. The following method has been proposed for the calculation of
these emissions.
USEPA Subpart W Approach:
Step 1: For each pump calculate the total blowdown volume in cubic feet (including, but not
limited to, pipelines, compressors and vessels) between isolation valves.
Step 2: Retain logs of the number of blowdowns annually.
Step 3: Calculate emissions using Equation W-24
Mass [subscript c, i] = N * V [subscript v] * R [subscript c] * GHG [subscript i] * 10^-3

Where:
1575
Mass [subscript c, i] = annual EOR injection gas venting emissions in metric tons at critical
conditions c from blowdowns
N = number of blowdowns for the equipment in the reporting year.
V [subscript v] = total volume in cubic feet of blowdown equipment chambers (including, but
not limited to, pipelines, compressors, manifolds and vessels) between isolation valves
R [subscript c] = density of critical phase EOR injection gas in kg/ft^3. Use an appropriate
standard method published by a consensus-based standards organization to determine density of
super critical EOR injection gas.
GHG [subscript i] = mass fraction of GHG [subscript i] in critical phase injection gas.
modification.
Response: EPA agrees with the comment and in todays final rule, EPA has retained the
equations to calculate EOR injection pump blowdown emissions as in the supplementary rule
proposal.
Organization: Commenter:
Section 98.233(w) EOR injection pump blowdown. Because carbon dioxide is the only
compound used for EOR that exists in a supercritical state, API believes this requirement should
be refocused. Further, the physical properties of carbon dioxide, while in the supercritical state,
are very difficult to determine outside of controlled laboratory conditions. API recommends the
following changes:
Section 98.233 (w) EOR injection pump blowdown.
Estimate carbon dioxide pump blowdown emissions as follows:
(1) Estimate the total volume in cubic feet (including, but not limited to, pipelines, compressors
and vessels) between isolation valves.
(2) Retain logs of the number of blowdowns per reporting period.
(3) Use Equation W-24 of this section to estimate the total annual venting emissions:
1576
Mass [subscript CO2] = N * V [subscript v] * R * 10^-3 (Eq. W-24)

Where:
Mass [subscript CO2] = Annual EOR injection gas venting emissions in metric tons at critical
conditions from blowdowns.
V [subscript v] = Total estimated volume in cubic feet of blowdown equipment chambers
(including, but not limited to, pipelines, compressors, manifolds and vessels) between isolation
valves.
R = Estimated density of critical phase carbon dioxide EOR injection gas in kg/ft^3. Retain
records of carbon dioxide density value.
Response: EPA disagrees with the commenter. The commenter has suggested the removal of the
GHG fraction in the CO2 stream being used for EOR. EPA agrees that the CO2 coming into the
EOR operations from a pipeline is typically almost all CO2. However, often, EOR operations
recycle the CO2 stream produced from the EOR operations back into the CO2 injection wells
without the separation of produced hydrocarbons. Hence, the CO2 stream at different points in
the EOR system will have differing levels of CO2 and EPA intends to gather that information.
Also, EPA has reduced the sampling burden from quarterly to annual and allows the use of
industry standard practices to determine density of super critical CO2. EPA, has therefore,
retained the GHG fraction calculation in todays final rule.
1577
13.2.11
DISSOLVED CO2 CALCULATIONS

98.233 (x) (y) These sections require quarterly samples be taken to determine dissolved CO2 in
produced fluids. Is there any provision to allow engineering estimates or less frequent testing (if
consistent values are obtained)
Response:In regards to the quarterly sampling of hydrocarbon liquids at EOR facilities, please
see response to EPA-HQ-OAR-2009-0923-0582-40 for further details.
In todays final rule, EPA does not require the monitoring of CO2 from produced water at any
onshore petroleum and natural gas production operations. For further clarification, please see the
98.233(x)-(y):
Hydrocarbon liquids dissolved CO2 and produced water dissolved CO2: The vast majority of
non-Enhanced Oil Recovery ("EOR") production fields do not have enough CO2 in their
hydrocarbon liquids and produced water to have more than a de minimis environmental impact.
We request that EPA state that this section applies only to hydrocarbon liquids and produced
water in CO2 injection EOR production fields.
Response: Todays final rule provides the necessary clarification to limit sampling of
hydrocarbon liquids to CO2 EOR facilities only. With regard to the sampling of hydrocarbon
liquids at CO2 EOR facilities, please see the responses to EPA-HQ-OAR-2009-0923-0582-40
and EPA-HQ-OAR-2009-0923-1011-21.

1578
In addition, 98.233(x) and (y) require quarterly samples at each hydrocarbon and produced water
tank to determine the concentration of dissolved CO2. With an average of 4 tanks/battery (2
hydrocarbon and 2 produced water) this cost would have a significant financial impact and
would be unnecessary. The same reputable laboratory referenced above has quoted
approximately $400 to analyze the CO2 in produced water and $700 to determine CO2 in
hydrocarbon liquids; and again, this is the cost for the laboratory analysis/sample ONLY, not the
time, labor or equipment to take the samples. Also, as stated above, the composition of the gas
would not vary significantly from battery to battery so quarterly sampling would be redundant.
The cost for the laboratory analysis alone to determine the CO2 retained in the hydrocarbon and
produced water tanks in Pioneer's Permian Basin operations only would be approximately $15.8
million. This will add millions of dollars for compliance, clearly not taken into account in EPA's
cost impact analysis. Further, these two provisions are nonsensical and the data would be
meaningless. Any CO2 or gaseous compounds would have been processed and contained prior to
reaching the storage tanks. Further, as the contents of the tanks are no longer under pressure, any
remaining CO2 would dissipate immediately so the samples taken at some point afterwards
would not be valid . Any amount left in tank would be negligible.
Pioneer requests that 98.233(x) and (y) be deleted from the Subpart. If these sections remain in
the final rule, Pioneer requests that an annual representative sample be taken from one oil and
one produced water tank in each basin, or alternatively in the reporting unit (per {(reporting unit"
recommendation in point 1). Further, section (y) states that sampling is not necessary if EOR
operations route produced water from separation directly to re-injection. As a second alternative,
if (y) remains in the final rule, Pioneer requests that produced water which is routed directly
from the tank to a salt water injection or disposal well be exempt as well. Lastly, another
potential approach is every time a well is completed, oil, water and gas samples could be taken at
that time and this data provided to the EPA. This would be an on-going obligation and not
limited to the annual compliance time periods.
Response: EPA agrees with the comment on sampling of hydrocarbon liquids. With regard to
the annual sampling of hydrocarbon liquids, please see the responses to EPA-HQ-OAR-2009
0923-0582-40, and EPA-HQ-OAR-2009-0923-1011-21.
please see the response to EPA-HQ-OAR-2009-0923-1151-129. These changes to todays final
rule will mitigate any cost concerns on the reporting of emissions from EOR facilities.

Hydrocarbon Liquids Dissolved CO2 & Produced Water Dissolved CO2 ( 98.233(x)-(y))
The proposed rule requires quarterly samples from each affected production site to determine the
1579
amount of CO2 retained in hydrocarbon liquids after flashing as well as in produced water at
atmospheric pressure.
We believe this requirement will not provide any valuable information because the CO2 that
would be measured is the amount of residual CO2 that is entrained in produced water or
hydrocarbon liquids and is not emitted to the atmosphere.
The quarterly sampling alone of these two fluid streams at each production site represents
approximately 60% of Devons costs.
Therefore, EPA should remove this requirement from the reporting rule.
Response: EPA agrees with the comment on sampling of hydrocarbon liquids. With regard to
the annual sampling of hydrocarbon liquids, please see the response to EPA-HQ-OAR-2009
0923-0582-40, EPA-HQ-OAR-2009-0923-1011-21, and EPA-HQ-OAR-2009-0923-1151-128.
Hydrocarbon liquids dissolved CO2 & Produced water dissolved CO2
Section 98.233(x) & (y): The vast majority of non-Enhanced Oil Recovery (EOR) production
fields do not have enough CO2 in their hydrocarbon liquids and produced water to have more
than a de minimis environmental impact. In addition, based on the placement of these sections in
the rule, it appears that EPA intended that Section 98.233(x) and (y) apply only to EOR. IPAMS
requests that EPA state explicitly that these sections apply only to hydrocarbon liquids and
produced water in EOR production fields.
Response:Todays final rule provides the necessary clarification to limit sampling of
hydrocarbon liquids to CO2 EOR facilities only. With regard to the sampling of hydrocarbon
liquids at CO2 EOR facilities, please see the responses to EPA-HQ-OAR-2009-0923-0582-40
and EPA-HQ-OAR-2009-0923-1011-21.
13.2.11.1
HYDROCARBON LIQUIDS
1580

Organization:
Commenter: S. Lampa
In addition, with regard to the gas industry, I am concerned about the impact that fracking will
have on my community's drinking water. From what I understand, in a worst case scenario, the
petroleum distillates used in a single well could contain enough benzene to contaminate more
than 100 billion gallons of drinking water to unsafe levels, according to drilling company
disclosures in New York State and published studies. (NYDEC DSGEIS 2009, Pagnotto 1961)
That is more than 10 times as much water as the state of New York uses in a single day.
(NYDECDSGEIS 2009)Fracking has already been linked to drinking water contamination and
property damage in Colorado, Ohio, Pennsylvania, Wyoming and other states. (Lustgarten
2008a, 2008b).Clearly, more transparency is needed by the drilling companies. It is important
that the Environmental Protection Agency use its existing authority to determine whether
companies are using diesel and enforce permit requirements.
Response: EPA has reviewed the commenters remarks. However, determining whether
companies are using diesel and enforcing permit requirements is beyond the scope of this rule.
6. Hydrocarbon Liquids Dissolved CO2
Source Description: Enhanced Oil Recovery operations that use injection of CO2 produce
petroleum that has significant amount of dissolved CO2. Although this CO2 is usually separated
from the petroleum product and re-injected, the liquid portion of petroleum still contains
dissolved CO2 which is released in a storage tank as the CO2 flashes out of the liquid
hydrocarbons. Losses will continue during transportation and processing stages.
USEPA Approach: The re-proposed subpart W contains the following method for the
determination of CO2 emissions.
Step 1: Determine the amount of CO2 retained in hydrocarbon liquids after flashing in tankage at
standard conditions. Quarterly samples must be taken according to methods set forth in
SECTION 98.234(b) to determine retention of Co2 in the hydrocarbon liquids immediately
downstream of the storage tank.
Step 2: Estimate emissions using Equation W-25 of this section.
1581
Mass [subscript s, CO2] = S [subscript hl] * V [subscript hl] (Eq W-25)

Where:
Mass [subscript s, CO2] = annual CO2 emissions from CO2 retained in hydrocarbon liquids
beyond tankage, in metric tons
S [subscript hl] = amount of CO2 retained in hydrocarbon liquids in metric tons per barrel, under
standard conditions
V [subscript hl] = total volume of hydrocarbon liquids produced in barrels in the reporting year
modification.
Response: EPA disagrees that EPA should use the same approach without an modification. In
todays final rule EPA has made some modification to the applicability and monitoring methods
for this source. Please see responses to EPA-HQ-OAR-2009-0923-1011-21 and EPA-HQ-OAR

XXI. Proposed Rule Section 98.233(x) Hydrocarbon liquids dissolved CO2. El Paso urges
EPA to limit the applicability of the requirements for dissolved CO2 specifically to onshore oil
and gas operations that utilize high CO2 content streams for enhanced oil production.
Response: In the April 2010 proposed rule found in docket (EPA-HQ-OAR-2009-0923-0002),
EPA did not intend to require sampling of hydrocarbon liquids at facilities other than at EOR
facilities. Todays final rule provides the necessary clarification to limit sampling of hydrocarbon
liquids at CO2 EOR facilities only.
1582

XXI. Section 98.233(x) Hydrocarbon liquids dissolved CO2
The proposed rule requires determining the amount of CO2 retained in hydrocarbon liquids and
produced water after flashing in tanks. The rule requires collecting these samples once each
quarter for every produced water or hydrocarbon tank.
This source has the highest cost per metric tons of emissions based on El Pasos cost estimates.
El Paso urges EPA to limit the applicability of the requirements for dissolved CO2 specifically to
onshore oil and gas operations that utilize high CO2 content streams for enhanced oil production.
This is the only operation for which this emission source makes sense and this is clearly the
intent for this emission sources as discussed in the TSD. The rule language should make this
distinction and only be applicable to EOR operations.
Response: EPA agrees with the comment. With regard to the sampling of hydrocarbon liquids at
CO2 EOR facilities, please see the response to EPA-HQ-OAR-2009-0923-1011-21.
98.233(x) Hydrocarbon liquids dissolved CO2
a.) We request EPA clarify that this paragraph applies only to hydrocarbon liquids production
associated with CO2 EOR operations. On page 52 of the TSD, EPA states Onshore petroleum
production that uses EOR with CO2 injection results in the production of petroleum with
significant amounts of CO2 dissolved in it. Limiting this paragraph to only hydrocarbon liquids
production associated with CO2 EOR operations, as it appears to be EPAs intent, will
significantly reduce the resource and analytical burden.
b.) This paragraph requires the quarterly sampling of liquid downstream from the tank for
dissolved carbon dioxide. In the case of the North Slope facilities, this will prove to be a
burdensome exercise without any benefit. Only a very small fraction of the oil produced by our
North Slope facilities is routed to storage tanks. This is done only to divert oil for non-routine
process reasons or because the oil is off-specification (slop oil) and the amount so routed is well
under 10%. Indeed, E&P Tanks will quantify under 98.233(j) any CO2 liberated from these
tanks. We thus request that this requirement not apply to our North Slope facilities.
Response: EPA agrees with the comment about hydrocarbon liquids at CO2 EOR facilities;
Please see the response to EPA-HQ-OAR-2009-0923-1011-21 for further details. If the North
1583
Slope facilities do not have CO2 EOR facilities, then the section of sampling of hydrocarbon
liquids does not apply to them.

Paragraphs (x)(1) and (y)(1), referring to hydrocarbon liquids and produced water dissolved in
CO2 requires quarterly sampling to determine retention of CO2 in each produced fluid. As stated
earlier, the composition of these streams does not fluctuate frequently enough to warrant
quarterly sampling. We propose every other year or annual sampling to satisfy the CO2 retained
in each fluid in equations W-25 and W-26.
Response: EPA agrees with the comment about annual sampling of hydrocarbon liquids. EPA
has determined that the quarterly sampling of hydrocarbon liquids at EOR facilities is unduly
burdensome without providing any substantial improvement in data quality. Hence, in todays
final rule EPA requires annual sampling of hydrocarbon liquids at CO2 EOR facilities.
Commenter:
Source Type
Hydrocarbon liquids dissolved CO2
Regulatory reference for calculation/ monitoring requirements
40 CFR 98.233(x)
1. Amount of CO2 retained in hydrocarbon liquids in metric tons per barrel, under standard
conditions.
2. Total volume of hydrocarbon liquids produced in barrels in the reporting year.
Comment:
It seems the EPA is assuming that the entrained CO2 will eventually be released, but it is unclear
why this assumption is being made. Further, it seems that the level of effort is not commensurate
with the amount of CO2 even the EPA expects to be emitted from this source. Also, this source
is not addressed in The Climate Registrys protocol for Oil and Natural Gas GHG reporting,
which is largely a more-inclusive program than EPA. If TCR did not address this as a source of
emissions, it is possible that this source is not expected to be a significant emitter of GHGs.
1584
Response: EPA wants to determine the amount of CO2 entrained in hydrocarbon liquids leaving
an EOR facility. Please see the response to EPA-HQ-OAR-2009-0923-1011-21. This is required
to determine the net balance of CO2 in an EOR system nationwide. Whereas EPA has tried to
adapt and conform with other programs, todays final rule is tailored to gather sufficient and
relevant data to inform future policy and therefore EPA has determined that this data is necessary
in order to do so.

Commenter:
Comment on 98.233(x)(1):
Regarding the requirement for quarterly sampling, WBIH requests a one-time sample
requirement. A sample is taken when the well begins production and that analysis will be used to
calculate emissions using direct measurement or E&P Tanks as outlined in previous comments
for 98.2330) and in the general comments (75 FR 18612). Quarterly sampling is not required for
current air quality compliance conditions and would be extremely burdensome and costly, which
EPA has likely significantly underestimated the cost involved with quarterly sampling
Response: EPA has determined that annual sampling of hydrocarbon liquids from EOR facilities
is sufficient for the level of data quality being sought. Please see the response to EPA-HQ-OAR
2009-0923-1058-13. However, a one-time sample is not deemed sufficient. The level of CO2 in
the reservoir and dissolved in the produced hydrocarbons varies significantly over the entire
EOR cycle. Hence it is important to monitor CO2 levels on an annual basis. Also, EOR
operations are not as extensive as conventional production operations and EPA expects this
reporting burden to be reasonable.
P. Section 98.233(x) Hydrocarbon liquids dissolved CO2
In the Technical Support Document (TSD), dissolved CO2 in hydrocarbon liquids is discussed in
relation to CO2 EOR operations. However, the proposed rule applies the category to all onshore
production. If EPA does retain this source category in the final rule, it should be restricted to just
CO2 EOR operations.
API believes the potential GHG emissions captured from this source category do not justify the
estimated costs of some $225.8 million per year. Based on a process simulation conducted by
one API member company, using an average of 4 condensate streams that had CO2 analyses at
1585
pressure in a Wyoming field, a residual CO2 in the liquid of 12.065 lbs/1,000 Bbls at 60 F
was determined. Applying this 12.065 lbs/1,000 Bbls to the total U.S. condensate production of
some 291.8 MM Bbls (EIA 2008) results in a total dissolved CO2 of just slightly under 1,600
metric tonnes. Applying this same result to the total US crude production of some 1.5 trillion
Bbls results in a total dissolved CO2 of about 10,600 metric tonnes.
The following table depicts the outcome of the process simulation:
[Table] Hydrocarbon CO2 Simulation
Based on these low emissions, which combined equate to a cost in excess of $18,000 per tonne
to comply with the proposed rule, API concludes that this source category is not justified. API
requests that this source be excluded from the reporting rule due to its insignificant contribution
to GHG emissions.
Response: EPA agrees with API that this source should be limited to CO2 EOR operations only.
Please see response to EPA-HQ-OAR-2009-0923-1011-21.
EPA notes the calculations on condensate CO2 emissions, but would like to point out that the
saturation of CO2 in hydrocarbons could be much higher in EOR operations. Overall, EPA
1586
intends to characterize CO2 dissolved in hydrocarbon liquids leaving CO2 EOR facilities and has
limited the source to EOR operations only, which reduces the burden to report.
39. Section 98.233(x) Hydrocarbon liquids dissolved CO2. The rule should state that this applies
only to produced water associated with EOR operations. As currently written, ALL hydrocarbon
liquids will require sampling for CO2. Additional details on this source type are provided in
Sections VI and VII.
Response: EPA agrees with the comment about hydrocarbon liquids at CO2 EOR facilities.
Please see the response to EPA-HQ-OAR-2009-0923-1011-21 for further details.
response to EPA-HQ-OAR-2009-0923-1151-129

S. Section 98.233(x) Hydrocarbon liquids dissolved CO2
If this source type is not eliminated from the final rule, it should state that this applies only to
produced water associated with CO2 EOR operations. As currently written, ALL hydrocarbon
liquids will require quarterly sampling and analysis for dissolved CO2. Similar to the GHG
volumetric emissions and Produced Water Dissolved CO2, BP requests that the EPA allow for
annual and representative samples. This change will not disrupt or misrepresent the reported
emissions; however, it will greatly reduce manpower and cost requirements to gather this
information. See comment #13 under EPA solicited comments for additional background and
context for this recommendation.
Response: : EPA agrees with the comment about annual sampling of hydrocarbon liquids at CO2
EOR facilities. Please see the response to EPA-HQ-OAR-2009-0923-1011-21 and EPA-HQ
1587

M. Section 98.233(x) Hydrocarbon liquids dissolved CO2
In the Technical Support Document (TSD), dissolved CO2 in hydrocarbon liquids is discussed
only in relation to CO2 EOR operations. However, the proposed rule applies the category to all
onshore production. If EPA does retain this source category in the final rule, it should be
restricted to just CO2 EOR operations rather than the broader industry.
BP believes the potential GHG emissions captured from this source category do not justify the
costs of quarterly sampling of every hydrocarbon tank in the oil and gas production industry.
Based on a HYSYS simulation using an average of 4 condensate streams that had CO2 analyses
at pressure in a Wyoming field, a residual CO2 in the liquid of 12.065 lbs/1,000 Bbls at 60 F
was determined. Applying this 12.065 lbs/1,000 Bbls to the total U.S. condensate production of
some 291.8 MM Bbls (EIA 2008) results in a total dissolved CO2 of just slightly under 1,600
metric tonnes. Applying this same result to the total US crude production of some 1.5 trillion
Bbls results in a total dissolved CO2 of about 10,600 metric tonnes.
The following table depicts the outcome of the HYSYS simulation:
[Table] Hydrocarbon CO2 Simulation
1588
Based on the low emissions coupled with the high cost of quarterly sampling and analysis the
cost per tonne covered is excessive and this source category should be dropped due to its
insignificant contribution to GHG emissions.
Response: EPA disagrees with the commenter about excluding the hydrocarbon liquids source
altogether from the rule. However, EPA has made changes to the requirements to limit the source
to EOR operations with simplified monitoring requirements. Please see responses to EPA-HQ
OAR-2009-0923-1011-21 and EPA-HQ-OAR-2009-0923-1058-13 for further details. In regards
to the condensate calculation, please see response to EPA-HQ-OAR-2009-0923-1151-128 for
further details.
13.2.11.2
PRODUCED WATER

Produced Water Dissolved CO2
Source Description: Enhanced Oil Recovery operations using injection of CO2 have produced
water that has dissolved CO2.
USEPA Approach:
1589
The re-proposed subpart W contains the following method for the determination of CO2
emissions in produced water
Step 1: Determine the amount of CO2 retained in produced water at STP conditions. Quarterly
samples must be taken according to methods set forth in SECTION 98.234(b) to determine
retention of CO2 in produced water immediately downstream of the separator where
hydrocarbon liquids and produced water are separated. Use the average of the quarterly analysis
for the reporting period.
Step 2: Estimate emissions using the Equation W-26 of this section.
Mass, CO2 = Spw * Vpw (Eq. W-26)
Where:
Mass, CO2 = Annual CO2 emissions from CO2 retained in produced water beyond tankage, in
metric tons.
Spw = Amount of CO2 retained in produced water in metric tons per barrel, under standard
conditions.
Vpw = Total volume of produced water produced in barrels in the reporting year.
EOR operations that route produced water from separation directly to re-injection into the
hydrocarbon reservoir in a closed loop system without any leakage to the atmosphere are exempt
from calculations in Step 1 and 2.
modification.
Response: In todays final rule, EPA does not require the monitoring of CO2 from produced
water at any onshore petroleum and natural gas production operations. For further clarification,

Commenter: Kim Dang
10. Produced water dissolved CO2. Kinder Morgan requests that this source be excluded from
the reporting rule due to its insignificant contribution of GHG emissions. EOR operations that
route produced water from separation directly to reinjection into the hydrocarbon reservoir in a
closed loop system without leakage to the atmosphere are exempt from SECTION 98.233(y).
This exemption from paragraph (y) should also apply to EOR operations that route produced
water from separation directly to disposal. As currently written, all produced water will require
sampling for CO2, which would be cost prohibitive.
1590

98.233(y) Produced water dissolved CO2
To emphasize API comment, this source should be excluded from the reporting rule due to its
insignificant contribution to GHG emissions. If EPA disagrees with this statement, EPA should
clarify that this paragraph applies only to produced water associated with CO2 EOR operations.
On page 53 of the TSD, EPA discusses the concern with dissolved CO2 in produced water in
relation to EOR operations. Limiting this paragraph to only produced water associated with CO2
EOR operations, as appears to be EPAs intent, will significantly reduce the resource and
analytical burden.

Commenter:
Source Type:
Produced water dissolved CO2
40 CFR 98.233(y)
1. Amount of CO2 retained in produced water in metric tons per barrel, under standard
conditions
2. Total volume of produced water produced in barrels in the reporting year
Comment:
It seems the EPA is assuming that the entrained CO2 will eventually be released, but it is unclear
1591
why this assumption is being made. Further, it seems that the level of effort is not commensurate
with the amount of CO2 even the EPA expects to be emitted from this source. Also, this source
is not addressed in The Climate Registrys protocol for Oil and Natural Gas GHG reporting,
which is largely a more-inclusive program than EPA. If TCR did not address this as a source of
emissions, it is possible that this source is not expected to be a significant emitter of GHGs.
Organization:
(98.233 (y)) includes Produced water dissolved CO2 emissions as a source of emissions to be
reported. On Page 18621 of the Federal Register, Item (2) Engineering First Principle Methods,
4th paragraph of the Subpart W preamble it states: "The supplemental proposed rulemaking does
not include emissions from tanks containing primarily water with the exception of transmission
station condensate tanks where dump valves are determined to be bypassing gas. Therefore, EPA
seeks comments on how to quantify emissions from tanks storing water without resulting in
additional reporting burden to the facilities." The source type Produced water dissolved CO2 in
the proposed rule and the statement in the preamble are contradictory. Please clarify the
intentions of Subpart W relating to water storage tanks

Commenter:
Comment on 98.233(y)(1):
Regarding the requirement for quarterly sampling, WBIH strongly requests a one-time sample
requirement. A sample is taken when the well begins production and that analysis will be used to
calculate emissions. Quarterly sampling is not required for current air quality compliance
conditions and would be extremely burdensome and costly, which EPA has likely significantly
underestimated the cost involved with quarterly sampling. For a cost example, in one production
field, there could be 1,000 wells, each with a production water pit. For this number of wells,
there would be 4,000 sampling and analysis events annually. This will cost approximately
$1,000,000 per year for analyses, with an additional cost for labor, just in one field
1592

Q. Section 98.233(y) Produced water dissolved CO2
In the Technical Support Document (TSD), dissolved CO2 in produced water is discussed in
relation to CO2 Enhanced Oil Recovery (EOR) operations. However, the proposed rule applies
the category to all onshore production. EOR and all produced water streams are managed for
minimal dissolved CO2 in produced water due to the severity of corrosion.
API believes the potential GHG emissions captured from this category do not justify the
estimated costs of some $323.1 million per year. Based on the maximum solubility of CO2 in
both pure water and sea water, API estimates the theoretical maximum dissolved CO2 in the 14.1
trillion barrels of produced water generated in 2008 (extracted from IHS provided production
data) as 4.7 MM metric tons, with an equilibrium saturation for fresh water of 2.1 grams/kg
(from online Engineering Toolbox) at 15 degrees C and less than 3.6 MM metric tons at an
equilibrium saturation for sea water of 1.55 grams/kg 232 if all produced water were CO2
saturated. In actual practice, most produced waters would not approach the saturation point and
actual dissolved CO2 would be a small fraction of this theoretical maximum.
The following tables depict APIs analysis of the theoretical CO2 content in produced water:
API believes that this source category should not be included in the final rule. The estimated
emissions based on the theoretical maximum CO2 are much larger than would be found in
practice. API requests that this source be excluded from the reporting rule due to its insignificant
contribution to GHG emissions.
232
"Solubility of Carbon Dioxide in Pure Water, Synthetic Sea Water, and Synthetic Sea Water Concentrates at -5
to 25 C. and 10- to 45-Atm. Pressure"; PAUL B. STEWART and PREM MUNJAL1; Sea Water Conversion
Laboratory and Department of Mechanical Engineering, University of California, Berkeley, Calif. 94720; Journal
of Chemical and Engineering Data, Vol. 15, No. 1, 1970.
1593
Response: EPA had intended produced water dissolved CO2 to be reported by EOR operations
only. However, EPA agrees with the calculation provided by API and has determined that the
emissions from produced water at EOR facilities may not significant enough to warrant reporting
at this point. Hence, todays rule does not require the monitoring of CO2 from produced water at
any onshore petroleum and natural gas production operations.
40. Section 98.233(y)(3) Produced water dissolved CO2. As demonstrated in Section VII.Q of
this document, API requests that this source be excluded from the reporting rule due to its
insignificant contribution of GHG emissions. EOR operations that route produced water from
separation directly to re-injection into the hydrocarbon reservoir in a closed loop system without
leakage to the atmosphere are exempt from paragraph (y). This exemption from paragraph (y)
should also apply to EOR operations that route produced water from separation directly to
disposal. As currently written, ALL produced water will require sampling for CO2 which would
be cost prohibitive. API recommends that the rule not require produced water dissolved CO2
sampling. Additional details on this source type are provided in Sections VI and VII.

- Section 98.232(c)(20) and Section 98.233(y): Similar to the Hydrocarbon Liquids Dissolved
CO2 category is the Produced Water Dissolved CO2 category. Again, EPAs Technical Support
Document only discusses this category in relation to CO2 EOR operations that route water
directly from separation to atmospheric storage tanks prior to recycle. However, the rule requires
quarterly sampling and analysis of the produced water directly following separation (presumed to
be at separator pressures) for all produced water streams. The proposed rule also requires
determination of the CO2 retained in the produced water at standard temperature and pressure
conditions presumably through use of a process model or laboratory GOR methodology,
although the rule is silent on the methodology. (Please note that the efficacy of this category is
discussed in another API comment specific to the source category provided in Section VII.Q.) As
with hydrocarbon tanks, this source requires measuring the CO2 in the produced water
immediately downstream of the separator, so an estimate of separators is used to evaluate the
costs. Assuming 308,000 separators require sampling, and applying a conservative estimated cost
of $475 per sample for the analysis, this yields a cost of $585 MM just for the required sampling
1594
and analysis. Adding a low estimate of 0.5 hours per tank for modeling of CO2 content at STP
(or an additional $50 per analysis for a laboratory GOR) yields an additional cost of $15.4 MM
for a total cost estimate of $600 MM. Note, this cost estimate is assumed to include produced
water from coal-bed methane operations.

25. CO2 Enhanced Oil Recovery Operations
We support EPAs proposed emission estimating method for CO2 released from petroleum and
produced water in storage tanks at Enhanced Oil Recovery (EOR) Operations based on
composition and volume.

T. Section 98.233(y)(3) Produced water dissolved CO2
If this source type is not eliminated from the final rule, it should state that this applies only to
produced water associated with CO2 EOR operations. As currently written, ALL produced water
will require sampling for CO2. Similar to the GHG volumetric emissions and Hydrocarbon
Liquids Dissolved CO2, BP requests that the EPA allow for annual and representative samples.
The proposed regulation requires that we sample and determine the CO2 content at the outlet of
each oil and water separation vessel. At many production facilities, plants have multiple
separators in series and multiple parallel separation trains. If the intent of this source category is
to capture produced water dissolved CO2 which may evolve in produced water tanks, only a
representative sample at the outlet of the last oil and water separation vessel should be required.
As reservoir fluids will be consistent throughout the year an annual sample is adequate to
estimate these emissions. If EPA chooses to keep this source category, BP requests that only an
annual and representative sample be required at the outlet of the last oil and water separator
upstream of the feed to the produced water tanks. This change will not disrupt or misrepresent
1595
the reported emissions; however, it will greatly reduce manpower and cost requirements to
gather this information.

O. Section 98.233(y) Produced water dissolved CO2
The only place that dissolved CO2 in produced water is discussed in the proposed rule
documents is in the TSD and only in relation to CO2 EOR operations however, the rule applies
the category to all onshore production. If EPA does retain this source category in the final rule, it
should be restricted to just CO2 EOR operations rather than the broader industry.
BP believes the potential GHG emissions captured from this category do not justify the costs
inherent in sampling and analyzing produced water from every separator in the oil and gas
production sector. Based on the maximum solubility of CO2 in both pure water and sea water we
estimate the theoretical maximum dissolved CO2 in the 14.1 trillion barrels of produced water
generated in 2008 (extracted from IHS provided production data) as less than 3.4 MM metric
tons with an equilibrium saturation for fresh water of 1.5 grams/kg (from online Engineering
Toolbox) at 20 degrees C and about 3 MM metric tons at an equilibrium saturation for sea water
of 1.325 grams/kg (from '"Solubility of Carbon Dioxide in Pure Water, Synthetic Sea Water, and
Synthetic Sea Water Concentrates at -5 to 25 C. and 10- to 45-Atm. Pressure"; PAUL B.
STEWART and PREM MUNJAL1; Sea Water Conversion Laboratory and Department of
Mechanical Engineering, University of California, Berkeley, Calif. 94720; Journal of Chemical
and Engineering Data, Vol. 15, No. 1, 1970) if all produced water were CO2 saturated. In actual
practice, most produced waters would not approach the saturation point and actual dissolved
CO2 would be a small fraction of this theoretical maximum.
The following tables depict this analysis of the theoretical CO2 content in produced water:
1596
BP believes that this source category should not be included in the final rule. The emissions
estimated based on the theoretical maximum CO2 are much larger than would be found in
practice and do not justify the projected costs associated with this source category. BP requests
that this source be excluded from the reporting rule due to its insignificant potential contribution
to GHG emissions.
13.2.12
PORTABLE EQUIPMENT COMBUSTION EMISSIONS

98.230 Definition of the source category.
We feel that the inclusion of portable non-self-propelled equipment (including but not limited to
well drilling and completion equipment, workover equipmentrented or contracted equipment)
with onshore petroleum and natural gas production creates a large tracking burden on reporting
entities in this category. Often times this equipment is transferred from reporting unit to
reporting unit depending on development needs. In addition rented equipment (i.e., well drilling,
completion and workover) location is not controlled by reporting companies. Rental companies
have multiple contracts in place with companies to frequently mobilize equipment to multiple
companies and reporting areas.
Response: EPA disagrees with the commenter on emissions from portable equipment. The
emissions contribution from portable equipment is significant enough to warrant data collection.
Please see the responses to EPA-HQ-OAR-2009-0923-1015-35 and EPA-HQ-OAR-2009-0923
1597

WCI recommends that SECTION 98.233(z) (Portable Equipment Combustion) be deleted and
the following paragraph be added to SECTION 98.232
(m) You must report the emissions of CO2, CH4, and N2O from each portable equipment
combustion source. For portable equipment combustion sources that combust field gas, you must
report under this subpart. For portable equipment combustion sources that combust fuels other
than field gas, you must report under subpart C of this part (General Stationary Combustion
Sources) using the methods required for stationary combustion sources.
Response: EPA disagrees with the commenter. However, EPA has clarified the overlaps
between Subparts C and W in regards to reporting of combustion emissions. For further details,
please see response to EPA-HQ-OAR-2009-0923-1060-27.
Commenter: Kim Dang
Clarification of When Portable Equipment Should Be Included. Kinder Morgan requests that
EPA eliminate its proposal to require reporting of emissions from portable non-self-propelled
equipment at onshore petroleum and natural gas production facilities. Including portable
equipment in Subpart W greatly magnifies the reporting burden for the onshore petroleum and
natural gas production sector, while only marginally increasing the coverage of the proposed
rule. Moreover, the TSD provides scant data or analysis to support the inclusion of portable
equipment in the Mandatory Reporting Rule; in fact, the TSD concedes that EPA does not even
have an estimate of disaggregated emissions from stationary and portable combustion devices in
the onshore petroleum and natural gas production sector.233
If EPA proceeds to require the inclusion of portable equipment in Subpart W, Kinder Morgan
requests that EPA provide clarification as to which equipment is covered. Specifically, proposed
40 C.F.R. SECTION 98.231(b) states that emissions from portable equipment that is stationed at
a wellhead for more than 30 days in a reporting year must be included when determining the
applicability of the rule. This requirement is overbroad, in that it sweeps in portable equipment
that is not stationed at production wellheads, as opposed to storage wellheads. EPA should revise
the Mandatory Reporting Rule to clarify that only portable equipment stationed at production
wellheads is to be included in applicability determinations and emission reports.
233
TSD, at 34.
1598
Response: EPA disagrees with the commenter. EPA has estimated that emissions from portable
equipment contribute to 45% of the total emissions from oil and gas production; please see
Portable Combustion Emissions memo under rulemaking docket EPA-HQ-OAR-2009-0923.
Hence it is not consistent with the broader objectives of the rule, which is to get reasonable
coverage of emissions to inform future policy, to drop portable equipment that are a significant
source of emissions from reporting.
EPA is not clear on the citation of the TSD footnote by the commenter. The very fact that EPA
does have disaggregated stationary and combustion emissions warrants the need to collect such
information. In todays final rule, EPA has provided a limited list of portable equipment that
needs to be reported for emissions, thus reducing the burden to report combustion emissions.
EPA has also provided an equipment threshold for external combustion equipment in onshore
production and natural gas distribution. Emissions from external combustion equipment equal to
or less than 5 mmbtu per hour are not required to be reported; however, equipment count by type
has to be reported. Finally, EPA has removed the 30-day clause from todays final rule. Please
Please refer to the Section 3, Summary of Comments and Responses of the preamble to
todays final rule for the response to these issues.

98.233(z): Calculate emissions from portable equipment using the Tier 1 methodology described
in subpart C of this part (General Stationary Fuel Combustion Sources).
OOC Comment: The rule should clarify that this subsection does not apply to offshore platforms.
Section 98.232 lists the GHGs to report. Subsections 98.232 (b) to (i) lists them by industry
segment, and only subsection (b) applies offshore. The remaining subsections, 98.232 (j) to (l),
do not include portable equipment combustion emissions. Thus, for offshore sources portable
equipment combustion emissions do not have to be reported. The only subsection where portable
equipment arguably have to be considered for offshore platform is subsection 98.231(b) dealing
with threshold calculation. However, that would result in use of Subpart C methodology to help
determine Subpart W threshold applicability. It is doubtful that this was the intended result.
Response: EPA has provided clarification in todays final rule that the Section 98.233(z)
indicated by the commenter applies to onshore production and natural gas distribution sources.
Offshore reporters shall report the emissions as calculated under BOEMRE compliance with 30
CFR 250.302 through 304.

1599

98.233(z) Portable equipment combustion emissions
98.233(z) seems to require annual reporting of emissions from portable emission units. This
conflicts with what appears to be EPAs intent, as noted on page 18619 of the rule preamble, to
use portable sources for applicability only. Portable sources should not be part of the annual
reports since those emissions are accounted for by the refiners who supply their fuel. We request
that EPA clarify this point by removing 98.233(z) from the rule before it is finalized.
Response: EPA agrees that the rule proposal may have caused some confusion in regard to the
portable combustion emissions requirement. In todays final rule, EPA has clarified that for
onshore production both portable and stationary combustion emissions must be included in both
the determination of threshold and for reporting of emissions. In addition, for natural gas
distribution stationary combustion emissions must be included in both the determination of
threshold and for reporting of emissions.
EPA disagrees with the commenter on exclusion of portable sources due to reporting of
refineries under Subpart MM. Although refineries will report total fuel supplied under Subpart
MM, EPA will not know where the fuels are being combusted to inform any combustion
equipment specific policy. Furthermore, onshore production combustion equipment typically use
field gas (or produced natural gas) that is not reported under Subpart MM or Subpart NN and
EPA intends to capture these emissions under Subpart W. For further details on double counting
of emissions, please see Section II.D of the preamble to The Final Mandatory GHG Reporting
Rule (Final MRR), (40 CFR part 98). Hence, in todays final rule, EPA has retained the
requirement for portable equipment emissions to be reported.

Commenter:
Source Type:
Combustion Equipment (including, but not limited to, engines, turbines, generators)
40 CFR 98.233(z)
1. For portable combustion equipment, use the Tier 1 methodology of Subpart C. The only
monitoring requirement for Tier 1 is annual fuel consumption.
2. For stationary combustion equipment, follow the requirements of Subpart C.
1600
There are several issues here: Does the 25,000 tonne threshold include all combustion equipment
in the basin for production facilities? Does this regulation bring all facilities reporting under
Subpart C into one larger facility under Subpart W? It is unclear how facilities required to report
under Subpart C by the standard definition of "facility" would also report under Subpart W.
Yates again requests clarification/removal of portable combustion equipment
Response: For onshore production, all portable and stationary combustion emissions in the
reporting basin must be included in the 25,000 MtCO2e threshold determination. However,
emissions from external combustion sources with a rated heat input capacity equal to or less than
5 mmBtu/hr are not included in the threshold determination (however, activity count has to be
provided). For threshold determination the reporter must combine emissions and follow the
requirements of 98.2, for facilities that contain any source category for which calculation
methods are provided.
In regards to reporting of combustion emissions, onshore production and LDCs, will report to
Subpart W under the common facility definition. All other segments in todays final rule will
report combustion emissions to Subpart C. For reporting of combustion emissions from LDCs,
For onshore petroleum and natural gas production, and natural gas distribution, all combustion
emissions monitoring and reporting requirements have been moved to Subpart W in todays final
rule. Hence, data reporting for the onshore petroleum and natural gas, and natural gas
distribution sectors combustion emissions starting 2011 shall be under Subpart W (note that for
year 2010 onshore production and natural gas distribution reporters have to comply with
requirements of Subpart C for combustion emissions, the switch is in force beginning 2011).
This change will avoid any conflicts with onshore production and natural gas distribution facility
definitions for combustion emissions between Subparts C and W.

Aka recommends that EPA remove the requirement of 98.236(f) which states that an operator
must report emissions separately for portable equipment for the following source types: drilling
rigs, dehydrators, compressors, electrical generators, steam boilers, and heaters. However, EPA
specifically exempts portable equipment from the definition of the stationary fuel combustion
source category in Subpart C of the rule. It is inconsistent and burdensome to require the
inclusion of portable sources for the petroleum and natural gas systems subject to Subpart W.
Response: EPA disagrees with the commenter with the exclusion of portable equipment from
todays final rule. However, EPA has provided an equipment threshold for external combustion
equipment in onshore production and natural gas distribution. For further clarification, please see
the responses to EPA-HQ-OAR-2009-0923-1024-23 and EPA-HQ-OAR-2009-0923-1060-27.
1601
With regard to control and maintenance of equipment by third-party contractors, please see the

Portable equipment should not be included - ie: drilling rigs
As an alternative to the request in point 5, if reporting emissions from all combustion sources in
Subpart W-applicable basins is required in the final rule, reporting emissions from portable
equipment should not be included. In Subpart C, portable equipment is specifically excluded;
however, Subpart W, 98.231 (b) states "By applying the threshold defined in 98.2{a){2}, you
must include combustion emissions from portable equipment that cannot move on roadways
under its own power or drive train and that is stationed at a wellhead for more than 30 days in a
reporting year, including drilling rigs, dehydrators, compressors, electrical generators, steam
boilers, and heaters." Complying with this requirement as proposed would be complex and
resource intensive, since this equipment is often moved from well to well or among operators.
Further, much of this equipment, especially compressors, drilling, completion, and workover
equipment, are contracted from a third party. When the operator does not own the equipment or
control the operation or maintenance of the equipment, they likely do not have the operational
data to perform the needed calculations and/or necessary records to certify the accuracy of the
GHG emissions data. Further, since this equipment is transient in nature, it does not accurately
represent continuous emissions at a facility. This would be impracticable to track and could
result in double-counting. Further, this section is unclear as to the 30-day requirement, if the
equipment is moved to various wells by the same operator in the same basin, or between basins,
etc ... Last, it is difficult to anticipate if a drilling rig, for example, will remain in a location
longer than 30 days.
Response: EPA disagrees with the commenter on not including portable equipment emissions
reporting. For further details on why portable equipment emissions are required to be reported,
please see response to EPA-HQ-OAR-2009-0923-1024-23.
EPA expects the reporting entity to collect relevant information from their contractors if need be
on portable equipment such as drilling rig emissions. The owner or operator that contracts a third
party has authority over the actions of that contractor and therefore the reporter must require the
contractor to provide the relevant information. For further details, please see response to EPA
HQ-OAR-2009-0923-1031-21 for further details. Finally, EPA has in todays final rule removed
the 30-day at wellhead clause to avoid practical issues with determining the time the portable
equipment is at the wellhead. Therefore, portable equipment located in non-reporting facilities
need not be tracked. Also, reporters will not have the additional burden of tracking the number of
days the portable equipment is operating in reporting facilities; reporters have to simply report
emissions from portable equipment located in the basin for the period it is in the basin.
1602
Furthermore, the equipment threshold for external combustion emissions will ensure that
equipment that is below this threshold does not have to be tracked.
Emissions from Portable / Non-self Propelled Equipment
Resolute Comments:
Resolute strongly opposes the requirement to report the emissions from non-stationary and
portable sources, especially well drilling rigs and ancillary equipment for well completions and
workovers. As is common practice in the industry, Resolute contracts with service companies to
drill and complete its wells. Complying with this requirement as proposed would be very
resource intensive and complex since this equipment is often moved from well to well, between
operators, and even between geologic basins. In addition, since this equipment is controlled and
maintained by third-party contractors, Resolute does not have access to the necessary
information or data to report their emissions, or to certify as to the quality of any data that
Resolute might be able to obtain. All other Clean Air Act programs establish applicability based
monitor the source's usage. No other industry sector is required to report contractor's emissions
under EPA's MRR, and it should not be required under Proposed Subpart W.
Response: EPA disagrees with the commenter on the exclusion of portable equipment emissions
reporting. For further details, please see response to EPA-HQ-OAR-2009-0923-1024-23. With
regard to control and maintenance of equipment by third-party contractors, please see the
EPA should not require the reporting of emissions from non-stationary and portable sources.
Response: EPA disagrees with the commenter on the exclusion of portable equipment emissions
reporting. For further details, please see response to EPA-HQ-OAR-2009-0923-1024-23.
1603
Section 98.233(z). Portable Equipment Combustion Emissions

EPA has indicated that they are especially interested in the significant sources of portable
combustion emissions of GHGs. To reduce the reporting burden for sources that are normally
owned and operated by a 3rd party, we recommend the requirement for this source be limited to
drill rigs, completion activities and compressors. In addition, to eliminate double counting of
emissions, this source category should exclude reporting emissions associated with burning fuels
subject to reporting under Subpart MM or burning NGLs subject to reporting under NN.
Response: In todays final rule, EPA has restricted the list of portable combustion equipment
required to report. However, EPA disagrees with limiting the list to just drill rigs and rental
compression. For further details, please see response to EPA-HQ-OAR-2009-0923-1060-27.
EPA disagrees with the exclusion of emissions associated with burning fuels under subpart MM
and burning NGLs under NN. For further details, please see the response to EPA-HQ-OAR
2009-0923-1042-26.
Recommended Leak Threshold
In order to provide a solution as to what defines a leak using either the imaging camera,
electronic gas screeners/detectors, infrared leak detectors or other devices, it is our opinion that
the leak should ultimately be determined on whether or not it can be measured with a high
volume sampler. This portable intrinsically safe measurement tool uses catalytic oxidation,
thermal conductivity hydrocarbon sensors to measure the gas concentration off the main sample
tube. A leak rate of 0.01 scfm can be measured which if leaking annually would amount to
approximately 5 Mcf or twenty-one dollars of gas at $4/Mcf. This is arguably small but at least a
threshold whereby the value of the gas can be determined and the economics of fixing the leak
known compared to the cost of repair.
Response: EPA disagrees with the commenter. Requiring direct measurement to comply with a
leak definition is onerous to reporters in terms of burden to report. Hence, EPA does not require
direct measurement to define a leak.

Organization: American Petroleum Institute, Karin Ritter
Section 98.233(z). Portable Equipment Combustion Emissions. EPA has indicated that they are
especially interested in the significant sources of portable combustion emissions of GHGs. To
reduce the reporting burden for sources that are normally owned and operated by a 3rd party,
1604
API recommends the requirement for this source be limited to drill rigs and rental compression.
In addition, to eliminate double counting of emissions, this source category should exclude
reporting emissions associated with burning fuels subject to reporting under Subpart MM or
burning NGLs subject to reporting under NN.
Response: In todays final rule, EPA has restricted the list of portable combustion equipment
required to report. However, EPA disagrees with limiting the list to just drill rigs and rental
compression. For further details, please see response to EPA-HQ-OAR-2009-0923-1060-27.
EPA disagrees with the exclusion of emissions associated with burning fuels under subpart MM
and burning NGLs under NN. For further information, please see the response to EPA-HQ
OAR-2009-0923-1042-26.
1605
VOLUME 14: DEFINITIONS AND BOUNDARIES
14.0
DEFINITIONS AND BOUNDARIES

Commenter: Kim Dang
SECTION 98.238 Definitions.
Except as provided below, all terms used in this subpart have the same meaning given in the
Clean Air Act and subpart A of this part.
Natural gas distribution facility means the distribution pipelines, metering stations, and
regulating stations that are operated by a Local Distribution Company (LDC) that is regulated as
a separate operating company by a public utility commission or that are operated as an
independent municipally-owned distribution system.
Offshore petroleum and natural gas production facility means each platform structure and all
associated equipment as defined in paragraph (a)(1) of this section. All production equipment
that is connected via causeways or walkways are one facility.
Onshore petroleum and natural gas production facility means all petroleum or natural gas
equipment associated with a all petroleum or natural gas production wells under common
ownership or common control by an onshore petroleum and natural gas production owner or
operator located in a single hydrocarbon basin as defined by the American Association of
Petroleum Geologists which is assigned a three digit Geologic Province Code. Where an
operating entity holds more than one permit in a basin, then all onshore petroleum and natural
gas production equipment relating to all permits in their name in the basin is one onshore
petroleum and natural gas production facility.]
Separator means a vessel in which streams of multiple phases are gravity separated into
individual streams of single phase.
Response: EPA disagrees with the commenters suggested change to the definition of onshore
petroleum and natural gas production facility to equipment associated with a well. EPA
considered and evaluated coverage of the onshore production segment at different reporting
thresholds and facility definitions, ranging from wellheads to fields to basins, arriving at a
conclusion that the basin definition was the most cost effective and reasonable coverage at the
25,000 tCO2e threshold. For further details, please see response to comment EPA-HQ-OAR
2009-0923-1167-1.
Commenter: Kim Dang
1606

Segmentation of the Petroleum and Natural Gas Sector. The proposed definitions of individual
segments of the petroleum and natural gas industry 234 contain unclear and overlapping references
to operations and equipment, and are impracticable for multipurpose or co-located facilities.
Many facilities owned or operated by companies in the petroleum and natural gas sector,
including Kinder Morgan, have equipment from multiple industry segments (processing,
production, etc.) located in the same physical space. Many facilities also serve more than one
purpose, such as both transmission and storage, or processing and transmission. The proposed
definitions do not provide guidance as to how facilities that serve or house multiple functions
should be categorized. Nor do the definitions provide the necessary certainty as to where the
physical boundaries of facilities should be drawn. This is a concern not just for multipurpose
facilities, but for facilities such as gas processing plants that are often surrounded by miles of
gathering pipelines and associated equipment owned and operated by multiple parties. Kinder
Morgans proposed changes to the definitions, which are provided in Appendix A, will clarify
the rule and increase the likelihood of accurate and consistent reporting.
As a supplement to the existing array of definitions, Kinder Morgan also supports INGAAs
suggestion to link the source category definitions to the primary North American Industrial
Classification System (NAICS) six-digit code that facilities would be required to report under the
revisions to Subpart A proposed on April 12, 2010. 235 This approach would provide a
straightforward way of classifying facilities, and would therefore help avoid misclassification of
facilities or equipment in the facilities and improve the ability of entities in the petroleum and
natural gas sector to submit consistent and accurate reports to EPA. If necessary, reporting
entities could provide additional information in their monitoring plan to justify the classification
of a facility whose NAICS code encompasses more than one of the facility types defined in the
proposed 40 C.F.R. SECTION 98.230(a)(1). The source category definitions should include
references to NAICS as follows:
- Offshore petroleum and natural gas production 21111
- Onshore petroleum and natural gas production - 211111
- Onshore natural gas processing plants 211112
- Onshore natural gas gathering compression 486210
- Onshore natural gas transmission compression 486210
- Underground natural gas storage 486210
234
40 C.F.R. SECTION 98.230(a).
235
Mandatory Reporting of Greenhouse Gases: Rule Amendment, 75 Fed. Reg. 18,455, 18,467 (Apr. 12, 2010)
(proposed amendments to 40 C.F.R. SECTION 98.3).
1607
- Liquefied natural gas (LNG) storage 486210

- LNG import and export equipment 486210
- Natural Gas Distribution 221210
Kinder Morgan suggests that a new source category for onshore natural gas gathering
compression be included in this the list, as more fully explained in Section II.B.5 of these
comments.
Response: In developing the petroleum and natural gas industry segment definitions EPA
considered the possibilities of multi-purpose facilities (e.g. transmission compressor stations that
alternate as gas storage compression) and operating sites with equipment from more than one
industry segment (e.g. transmission compression station with a gas processing straddle facility).
Because there are so many possible combinations of equipment that can serve multiple purposes
and operating locations with multiple industry segments represented, EPA concluded that it
would be intractable to define industry segments in such a way that only one industry segment
would be clear for every possible combination. EPA also considered that the emissions
estimation methodologies are the same for segments that have or share similar equipment as
other segments (e.g. processing, transmission, gas storage, LNG compressors). Todays final rule
requires the reporter to determine the industry segment for which the majority of emissions occur
and report all equipment within that facility for which there is a method defined. EPA recognizes
that some data may not be reported where the obvious industry segment (e.g. processing facility)
has one or a few producing wells inside the facility fence and the processing facility
owner/operator does not have enough production emissions in the basin surrounding the
processing facility to meet the reporting threshold under onshore production. EPA concludes that
such random missing data will not impact the overall inventorys ability to inform policy.
Where a piece of equipment serves a dual purpose, such as a transmission compressor operating
part time as a gas storage compressor, the rule requires this piece of equipment to be reported
under the majority use industry segment. EPA concluded that most of the potential dual-purpose
equipment will have the same emissions reporting requirements in either industry segment, so
reporting under only one majority use segment will avoid double counting.
EPA disagrees with the comment on the use of the term facility, as EPA has used the term
facility consistently across the entire MRR to collect necessary GHG data. Please see response
to comment EMAIL-0001-1 (comment also located in rulemaking memo Early Comment
Submissions in docket EPA-HQ-OAR-2009-0923). EPA considered the addition of NAICS
codes in the definition of industry segments, and rejected this because the codes are not
sufficiently unique to add differentiation to some industry segments. For example, one NAICS
code, 486210, applies to Onshore natural gas gathering compression, Onshore natural gas
transmission compression, Underground natural gas storage, Liquefied natural gas (LNG)
storage, and LNG import and export equipment. Please see response to comment EPA-HQ
OAR-2009-0923-1039-6.
1608

Commenter: Kim Dang
Gathering Facilities Should Report Separately and in the Same Manner as Natural Gas
Transmission Compression Facilities. EPAs proposal to designate gathering pipelines as part of
onshore petroleum and natural gas production and onshore natural gas processing plants 236 does
not conform to the structure of the industry. Gathering compression facilities and pipelines have
never been considered by the industry to be part of onshore production facilities or processing
plants. Nor has EPA grouped gathering compression facilities and pipelines together with these
facilities in its other CAA programs, such as NSPS. 237 Even setting these considerations aside, it
is difficult to see how a reporting entity that owns a processing plant or production well could
properly take gathering systems emissions into account, considering the complexity and multiple
ownership and operating arrangements of gathering pipeline networks. Gathering pipelines can
extend for miles around a processing facility or production area. Multiple different gathering
pipelines owned by multiple different entities are often connected to the same processing facility
or production area. Gathering pipelines and the associated compressor stations can often have the
ability to deliver to more than one processing facility. Thus, it would often be inaccurate and
infeasible to attribute emissions from a gathering pipeline system to a particular processing
facility. Indeed, it is simply impossible to lump these complex networks together with discrete
processing facilities and production areas in a way that is logical and can be consistently
implemented.
Furthermore, in many respects the same rationales that correctly led EPA to exclude petroleum
and gas pipeline segments from Subpart W 238 apply with equal or greater force to gathering
pipeline segments. Gathering pipelines are just as widely dispersed as transmission pipelines, are
smaller in diameter than transmission pipelines, and operate at lower pressure than transmission
pipelines. As with transmission pipelines, gathering pipelines are aggressively surveyed for
leaks. For example, those gathering pipelines that are under the jurisdiction of the Pipeline and
Hazardous Materials Safety Administration (PHMSA) are subject to federal regulations
imposing regular inspections and repair requirements similar to those that apply to natural gas
transmission pipelines. 239 There are also some 250,000 miles of gathering pipelines in service
around the U.S.; if reporting entities were required to conduct component counts and carry out
236
Proposed 40 C.F.R. SECTION 98.230(a)(2), (a)(3).
237
See 40 C.F.R. SECTION 60.630, 631 (defining natural gas processing plant to include the processing site, and
only those portions of the field gas gathering system that are actually located on the processing site).] or
NESHAPS. [Footnote 29: See 40 C.F.R. SECTION 63.761 (defining facility and natural gas processing plant
for NESHAPS for Oil and Natural Gas Production Facilities).
238
239
See 49 C.F.R. SECTION 192.706, 711. EPAs TSD for the proposed Subpart W identifies PHMSA regulations
as one reason why natural gas transmission pipeline segments should not be included in the Mandatory Reporting
Rule.
1609
optical leak detection surveys along the full length of these lines, the costs of Subpart W would
escalate far beyond what EPA has estimated.
Kinder Morgan recommends that the proposed rule be revised to treat gathering compression and
pipelines similarly to transmission compression and pipelines. Under this approach, gathering
pipeline compressor and booster stations would be defined as a separate category of facility in
proposed 40 C.F.R. SECTION 98.230, rather than being associated with production or
processing facilities. Combustion, vented, and fugitive emissions from boosting and gathering
stations that are above the 25,000 metric tons CO2-e per year threshold would be reported in the
same manner as emissions from transmission compressor stations. Gathering pipeline segments,
or at least those subject to PHMSA regulations, would be excluded from reporting. This
approach would provide reporting of significant emissions from gathering pipeline systems,
while eliminating the inconsistencies and inaccuracy that would surely result if gathering
compression or pipelines were reported as part of the production or processing sectors.
Response: Todays final rule does not require reporting of emissions from gathering and
boosting segment of the industry. For further details, please see Section II.F of the preamble to
todays final rule.

Commenter: Kim Dang
Clarification of Coverage of CO2 Transmission Pipelines. Our review of the rule suggests that
EPA did not intend to require reporting of emissions from pipeline facilities that exclusively
transport CO2. Nevertheless, Kinder Morgan is concerned that its CO2 transmission pipelines,
which transport CO2 extracted from natural deposits to various EOR operations, could be
considered part of an EOR operation for purposes of Subpart W, 240 and wishes to avoid
confusion in the final rule. CO2 pipelines are, of course, distinct from natural gas transmission
pipelines, and are not specifically mentioned as part of the petroleum and natural gas production
sector in either the preamble to the proposed rule or the accompanying TSD.
As a policy matter, Kinder Morgan believes that CO2 transmission pipelines should be treated
under Subpart W in a similar manner to natural gas transmission pipelines. Any facility along the
pipeline that exceeds the 25,000 metric tons CO2-e per year reporting threshold would be subject
to reporting, but the segments of pipe between facilities would be excluded from reporting.
Excluding CO2 transmission pipeline segments from Subpart W would be appropriate, given that
these pipelines present the same measurement issues that justify the exclusion of oil and natural
gas pipelines from Subpart W (i.e., dispersed and relatively minor emissions). PHMSA
regulations also require at least 26 surveys of surface conditions along pipeline rights-of-way
240
See Proposed 40 C.F.R. SECTION 98.230(a)(2) (including all EOR operations using CO2 in the definition of
onshore petroleum and natural gas production).
1610
each year, 241 and Kinder Morgan takes the additional precautionary measure of continuously
monitoring for leaks in its CO2 pipelines using a Supervisory Control and Data Acquisition
(SCADA) system. Accordingly, Kinder Morgan requests that EPA clarify that CO2 transmission
pipeline segments between facilities are excluded from reporting under Subpart W.
Response: Todays final rule does not include methodologies for emissions quantification or
reporting of transmission gas pipelines between production, compression or processing facilities.
This applies equally to CO2 pipelines. Todays final rule, Section 98.232 - GHGs to report,
specifies what GHG emissions to report, and if an emission source is not listed in this section of
the rule, it is not required to be reported. CO2 transmission pipelines, which transport CO2
extracted from natural deposits to various EOR operations, are not considered part of an EOR
operation for purposes of Subpart W.
Commenter: Kim Dang
EPA should revise its source category definitions to avoid confusion and uncertainty, especially
with respect to multipurpose and co-located facilities. Using primary NAICS codes to classify
facilities is recommended.
Response:
EPA does not agree to revise source categories or facility definitions for multipurpose or co
located facilities, or to include NAICS codes as part of facility definitions. Please see response
to comment number EPA-HQ-OAR-2009-0923-1024-14 for more information.
14.1
ONSHORE PRODUCTION FACILITY DEFINITION

B. Onshore Production: Basin-Level Definition of "Facility"
Under the proposed Subpart W, a single "onshore petroleum and natural gas production facility"
means "all petroleum or natural gas equipment associated with all petroleum or natural gas
production wells under common ownership or common control by an onshore petroleum and
natural gas production owner or operator located in a single hydrocarbon basin as defined by the
American Association of Petroleum Geologists ["AAPG"] which is assigned a three-digit
Geologic Province Code." 40 C.F.R. 98.238 (emphasis supplied), 75 Fed. Reg. 18608, 18647
(April 12,2010). Significantly, "[w]here an operating entity holds more than one permit in a
241
See 49 C.F.R. SECTION 195.412.

1611
basin, then all onshore petroleum and natural gas production equipment relating to all permits in
[that entity's] name in the basin is one onshore petroleum and natural gas production facility." Jd
(emphasis supplied). In the Preamble to its proposal, USEPA explains that this approach was
selected to capture "approximately 81 percent of emissions from onshore petroleum and natural
gas production." 75 Fed. Reg. at 18615.
IOGA-WV strongly opposes this approach, which plainly (indeed, intentionally) disregards the
most obvious and logical definition of a onshore production facility-the well pad, which may
contain one or more wells and the attendant separation and storage facilities and associated
equipment. Such an approach would be consistent with permitting practices for these facilities
common under federal and state air programs. Proposed Subpart W, by contrast, artificially and
dramatically broadens the definition of "facility" and effectively requires the industry sector to
conduct additional monitoring and to adhere to a reporting threshold far below the 25,000 tpy
CO2e threshold that applies to other regulated industries. Obviously, this will impose
substantially greater costs on the oil and gas industry as compared to other industry sectors,
while producing only marginal environmental benefit.242
Response: EPA does not agree that the onshore facility definition is contrary to the CAA
requirements. Please see Topic 1: GHG Reporting under Subpart W and the Consolidated
Appropriations Act in Volume 9 of the Response to Comments and the response to EPA-HQ
OAR-2009-0923-1044-1 for further details. EPA also disagrees with the commenter on the use
of the well pad facility definition. Please see EPA-HQ-OAR-2009-0923-1005-3 for further
details.

In conclusion, IOGA-WV fundamentally disagrees with USEPA's conclusions in the Preamble to
its proposed Subpart W that the GHG emissions from oil and gas production sector arc sufficient
to warrant imposing burdensome monitoring and reporting requirements for individual wells. See
75 Fed. Reg. at 18613-15. IOGA-WV harbors serious concerns about USEPA's proposed
adoption of a basin-level definition of an onshore production "facility" solely to capture small
sources of GHG emissions that otherwise would fall well below the 25,000 tpy CO2e
applicability threshold that represents a balance of the environmental benefits of the rule with the
burdens on small emitters. IOGA-WV believes that this approach would set a troubling
precedent that, if taken to its logical conclusion and applied to other industries, could justify
reporting for almost every small source of GHG emissions in the United States. Accordingly,
IOGA-WV urges USEPA to adopt a well pad definition of facility that is reflective of actual
operations. As noted in the preceding paragraph, while small emitters would be exempt from
242
I IOGA-WV refers USEPA to the statistics set fOlih in IPAA's comments suggesting that onshore petroleum and
natural gas production operations account for approximately 3.9% of the nation's GI--IG emissions.
1612
reporting obligations under this approach, USEIA has capabilities that will allow it to make a
reasonable estimate of those emissions attributable to onshore oil and gas production facilities.
Response: EPA disagrees with the commenter on the use of the well pad facility definition.
Please see EPA-HQ-OAR-2009-0923-1005-3 for further details. EPA has also determined that
only a small number of marginal operators in the country will be impacted by the rule. Please see
response to EPA-HQ-OAR-2009-0923-1005-7 and Section 5.2 of the Economic Impact
Analysis (EIA) for todays final rule found in docket (EPA-HQ-OAR-2009-0923)for further
details.
Commenter:

Facility Definitions
Industry and regulators currently have a very good understanding of what the term facility
means in the O&G upstream sector. Company representatives as well as state regulatory agencies
have been using this term to apply current CAA requirements across the country. Our
recommendation is that we rely on the current actively used definition of facility that EPA
change their proposed language to refer to a threshold reporting area or roll-up area
threshold to define a geographical area to which a threshold can be applied that is made up of
oil and gas upstream facilities. This would help alleviate industry concerns that the reporting rule
definition of facility might impact perception of the CAA definition.
For example, 98.238 Onshore petroleum and natural gas production facility would read:
Onshore petroleum and natural gas production reporting area means all petroleum or natural gas
facilities and equipment associated with all petroleum or natural gas production under common
operator located in a single hydrocarbon basin as defined
The term reporting area could then be referred to in the definition of the source category; i.e.,
98.230(2) Onshore petroleum and natural gas production: Onshore petroleum and natural gas
production reporting area means all facilities associated with wells the production of petroleum
or natural gas including but not limited to
Response: EPA does not agree with the commenter. EPA has used the term facility
consistently across the entire MRR to collect necessary GHG data. Also, all the requirements set
forth in Subpart A refer to the facility. The use of any other term will lead to a disconnect within
the MRR on the applicability of these overarching requirements. Furthermore, EPA does not see
any benefit in using a term other than facility. Hence, in todays final rule, EPA has retained the
usage of the term facility where appropriate. In todays final rule EPA has revised the term
facility for the purposes of Subparts W and separated it from the Subpart A requirements with
respect to natural gas distribution and onshore petroleum and natural gas production for clarity.
1613
EPA has also ensured that there is no double counting of emissions [for monitoring, reporting
and applicability threshold purposes] of co-located facilities in those source categories; please
Commenter:

If Upstream Production Sites are Aggregated
If EPA rejects important financial burden and CAA definition of facility arguments and retains
aggregation of upstream production sites, it should take reasonable steps to mitigate the adverse
financial burden to industry, especially very small operators. This would include:
Exempting small sites using well defined screening thresholds such as (1) natural gas
production sites with only wellhead fugitive emissions and produced water storage tank
emissions, (2) any natural gas production site with less than 3 MMSCFD of production and less
than 1 BOPD condensate production, and (3) oil stripper wells producing less than 10 BOPD.
Allowing the use of the API Compendium for calculations of emissions from upstream
production sites. (See Item 3 below).
Response: EPA does not agree with the commenter on the exemption of small production sites
based on their production level. Any such exclusion would necessitate defining the term small
site, which is not feasible for the same reason as defining a production facility site using
aggregation of equipment; please see Section 4.c. of the TSD for further details. However, to
reduce burden on the industry, EPA intends to develop a screening tool to assist in the
determination of which facilities are required to report under todays final rule; please see
Section II.F of the preamble to todays final rule for further details..
In addition, glycol dehydrators, onshore production storage tanks and blowdown vents below set
equipment thresholds can report emissions using a population emission factor in todays final
rule. However, those emission sources equal to or above the equipment threshold are required to
estimate emissions using more rigorous methodologies. For further information please see the
response to EPA-HQ-OAR-2009-0923-1011-39, EPA-HQ-OAR-2009-0923-1060-10, and EPA
HQ-OAR-2009-0923-1061-12 for glycol dehydrator, onshore production storage tanks and
blowdown vents, respectively. In addition, external fuel combustion emissions from portable and
stationary equipment, in onshore production and natural gas distribution, with heat input less
than or equal to 5 mmBtu/hr do not have to be reported; however, reporting of equipment count
by type is required. For further information, please see EPA-HQ-OAR-2009-0923-1024-23.
Further details are available in Section II.F of the preamble to todays final rule.
1614
EPA disagrees with only the use of the API Compendium emissions factors for calculations of
emissions. However, EPA has used the same Gas Research Institute Study Methane Emissions
from the Natural Gas Industry that the API Compendium uses with some modifications to
develop emissions factors for todays final rule. Please see rulemaking docket EPA-HQ-OAR
2009-0923 for memo Equipment-Level Population Emission Factors for Onshore Production
and response to EPA-HQ-OAR-2009-0923-1151-130 for further details.

EPA Must Clarify the Proposed Definitions of Industry Segments Subject to Subpart W
Like the American Petroleum Institute and other industry commenters in this docket, NGSA
believes that the definitions of individual segments of the oil and gas industry in the proposed 40
C.F.R. 98.230(a) are confusing, inconsistent with industry practice, and would lead to doublecounting of emissions from multipurpose or colocated facilities.
For example, it is not clear from the segment definitions whether a given gathering pipeline
should be considered part of an onshore petroleum and natural gas production facility or a
natural gas processing facility, because the definitions do not specify a physical boundary for
either production or processing. Similarly, other units (such as gas dehydrators) could be
considered part of more than one facility under the definitions proposed by EPA. Industry
practice and existing EPA regulations differ substantially from the proposed Subpart W by
clearly demarcating individual types of facilities with reference to a physical boundary, and by
excluding gathering pipelines outside the facility fence-line. Because the definitions in the
proposed 40 C.F.R. 98.230(a) are ambiguous and difficult to apply, they would lead to
inconsistent and potentially overlapping reports. These uncertainties also create a risk that NGSA
members could be held liable for misreporting, even when making a good faith effort to properly
apply the definitions in the rule.
In addition, the proposed definitions provide no guidance as to how facilities that serve or house
multiple functions should be categorized. In the natural gas industry, it is common for a single
site to include multiple facilities that perform different functions (production, processing,
compression, etc.). Under the proposed Subpart W, it is not clear to which industry segment such
facilities would correspond. Without further specifications on the treatment of such facilities, the
proposed Subpart W could lead to inconsistent reports or inadvertent misreporting of emissions.
To resolve these issues, NGSA recommends that EPA:
Draw on existing regulatory definitions of facilities where possible (such as the descriptions
provided in Subpart HH of EPAs NESHAP regulations for natural gas processing facilities, 40
C.F.R. 63.760). These definitions are well-understood by industry, and have proved workable in
other regulatory contexts;
1615
Create a separate facility category for gathering pipelines, consistent with the existing
Department of Transportation / Pipeline and Hazardous Materials Safety Administration
reporting;
Treat gathering/boosting compressor stations as individual facilities, consistent with current
definition and treatment of facilities under the CAA; and
Allow reporting entities to use primary North American Industrial Classification System
(NAICS) codes, and other appropriate and relevant data, to justify the classification of multi
purpose or co-located facilities.
Response: As explained in EPA-HQ-OAR-2009-0923-1044-1, EPA concluded that the
definitions contained in todays final rule are appropriate to the purposes the MRR and Section
114. To apply NESHAP definitions or other definitions does not meet the purposes of the
reporting rule. Such definitions are too restrictive and in application would exclude too many
GHG emissions.
Todays final rule does not include gathering lines and boosting stations and hence the issue of
gathering pipelines is not relevant. For further clarification, please see Section II.F of the
preamble. In the final rule EPA has revised the term facility in 98.238 for the purposes of
subpart W, and separated it from the Subpart A requirements with respect to natural gas
distribution and onshore petroleum and natural gas production for clarity. Please see the
preamble Section II.D.
EPA agrees there is necessity for clear requirements for co-located facilities and multipurpose
equipment. In todays final rule, EPA has provided further guidance on how reporters should
deal with co-located facilities that serve multiple functions and also on equipment that serve
multiple purposes. Please see the response to EPA-HQ-OAR-2009-0923-1024-14 for collocated
facility requirements and equipment that serve multiple purposes.
EPA disagrees with the commenter on the use of NAICS code system. For further details, please

Facility Definition
Most notably, we believe that use of the CAA denies EPA the authority to create a definition of a
facility that differs from that in the CAA. EPA proposes the following definition:
equipment associated with all petroleum or natural gas production wells under common
1616

petroleum and natural gas production facility.
Under this definition, for example, all wells under common ownership along the Gulf Coast of
Texas and Louisiana and deeply into the mainland of those states would be considered as one
facility. This would be analogous to proposing that every McDonalds restaurant in the State of
Texas should be considered as one facility because they have the same name and are franchised
from a common source.
Nothing in the CAA suggests that EPA can define an onshore petroleum and natural gas
production facility as broadly as it proposes. In reality, the only guidance provided to EPA in the
CAA resides in Section 112(n)(4)(A) where it states:
in the case of any oil or gas exploration or production well (with its associated equipment),
such emissions shall not be aggregated for any purpose
Response: See response to Comment Number EPA-HQ-OAR-2009-0923-1044-1. Additionally,
EPA finds the commenters reference to 112(n)(4)(A) unpersuasive. The reference to
aggregation contained therein relates exclusively to major source determinations for the purposes
of the control of hazardous air pollutants and further expressly limits the provision to any
purpose under this section., namely CAA Section 112. As pointed out in previous responses
to comments the reporting rule is not delimited by other CAA programs.
Furthermore, EPA is obligated to collect all relevant GHG emissions data from all the industrial
sectors to inform future policy making, including onshore production of the oil and gas industry.
On a national basis, GHG emissions from the oil and gas industry are second only to power
plants. And within the oil and gas industry, the onshore petroleum and natural gas sector
contributes over half of those total emissions. Hence, as the upstream oil and gas production
sector contributes such a large portion of total U.S. greenhouse gas emissions, EPA has
determined that it is necessary to collect an accurate amount of GHG emissions information from
this segment. In developing todays final rule, EPA considered various options in collecting the
information.
Corporate or company level reporting nationwide was an option EPA considered but did not
pursue since it was substantially more burdensome for small businesses than a basin facility
definition because companies are bigger in size than are facilities for the application of emissions
thresholds, i.e. companies would end up adding up emissions from their multiple facilities, in this
case basins.
EPA considered field level and wellhead level facility definitions as well. To meet the purposes
of the MRR and to gather a reasonably significant amount of GHG emissions information from
the sector using these facility definitions, would have required either drastically lowering or
eliminating the 25,000 ton threshold altogether. Defining the source category on a field or
1617
wellhead basis would require all-in reporting of emissions to gather any meaningful amount of
information such as with respect to those sources covered by Section 98.2(a)(1) of todays final
rule and would be substantially more burdensome than the basin-level definition because many
more facilities would have to report. Therefore, EPA has determined that defining the facility at
the basin level best manages the burden issue that the field level and wellhead level definitions
raise. A basin level definition balances the cost to report with the coverage of emissions. For a
wellpad level coverage of emissions see Section 5 of the TSD to todays final rule for further
details. For field level coverage and cost analysis see Section 5.1.4 of the EIA to todays final
rule.
Furthermore, only a limited number of small businesses will have to report under the basin level
definitions; please see response to EPA-HQ-OAR-2009-0923-1005-7 and Section 5.2 of the EIA
for further details. Conversely, a significant number of small businesses would potentially be
impacted if a field level or wellhead level definition were adopted because facilities are smaller
when using field level or wellhead level definition and no or lower than 25,000 metric tons CO2e
threshold. . In conclusion, EPA has the Congressional mandate to adequately collect meaningful
information from all segments of the economy including onshore production and considers
todays final rule to be the most cost-effective way to collect that information.

Facility Definition
Xcel Energy owns and operates natural gas systems that include production, processing,
transmission, storage, and distribution. For Subpart NN compliance, the entire system was
considered as an LDC distribution system because the pipeline distribution system operates
under a single tariff with oversight by the state Public Utilities Commissions or the Federal
Energy Regulatory Commission.
Reporting of CO2 emissions under Subpart NN will be straightforward and the resources for
reporting can be identified and scheduled. However, the definition of a "facility" in Subpart W
lacks clarity as it relates to the 25,000 mtCO2e applicability determination. The Preamble
acknowledges this lack of clarity (pg. 18613), but does not provide much guidance concerning
the relation to stationary combustion sources in Subpart C that are below the reporting threshold.
Xcel Energy recommends that stationary combustion sources that are excluded because they are
below 25,000 mtCO2e for Subpart C be excluded from Subpart W reporting.
Response: EPA does not agree with the commenter on the exclusion of combustion emissions
sources that are not included in Subpart C. In todays final rule, an LDC reporting to Subpart W
shall include all combustion emissions, less combustion emissions from equipment equal to or
below 5 mmBtu/hr rated heat input capacity, in the determination of threshold to report for
Subpart W, i.e. 25,000 mtCO2e (a count by type of equipment at or below the 5 mmBtu/hr
1618
equipment threshold has to be reported). Furthermore, starting 2011, LDCs are required to use
facility definition and monitoring methods for combustion emissions under Subpart W. Also,
starting 2011 all LDCs shall report combustion emissions under Subpart W.

Preamble Section II.(C.) on page 18613.El Paso requests the EPA to consider utilizing the
term reporting area under the definition of the source category for Onshore Petroleum and
Natural Gas Production.
Response: EPA does not agree with the use of the term reporting area instead of facility. For
further details, please see the response to EMAIL-0001-1 (comment also located in rulemaking

Proposed Rule Section 98.238 Definitions.Onshore petroleum and natural gas production
facilities should be renamed reporting areas for clarity and correctness.

Section 98.238 Definitions
Onshore petroleum and natural gas production facility
Consistent with our comments under Section II, we recommend the use of facility be replaced
with reporting area as follows:
Onshore petroleum and natural gas production facility reporting area means all petroleum or
1619
natural gas equipment associated with all petroleum or natural gas production wells under
common ownership or common control by an onshore petroleum and natural gas production
owner or operator located in a single hydrocarbon basin as defined by the American Association
of Petroleum Geologists which is assigned a three digit Geologic Province Code. Where an
petroleum and natural gas production facility reporting area.

98.233(g)(ii)(C): Gas well venting during unconventional well completions and workovers. The
term producing field in this subpart is not defined.
Response: EPA has defined the term field in todays final rule and a producing field simply
means a field that has any production during the reporting period. Please see Section 98.238 of
todays final rule.

It is unclear how facilities required to report under Subpart C by the standard definition of
"facility" would also report under Subpart W. PAW requests clarification/removal of portable
combustion equipment
Response: The EPA has made clarifications about any conflicts for combustion emissions
between Subparts C and W. For further details, please see the response to EPA-HQ-OAR-2009
0923-1060-27.

1620
EPA is proposing an expansive and novel definition of the term facility in this 2010 Proposal
that would sweep in all the miles of gas mains and customer service lines, city gate stations, and
(due to an unclear definition) potentially all customer meters across a state if they are within a
distribution system served by a single LDC.
Response: EPA does not agree with the commenter. EPA has clarified the definition for LDCs
in todays final rule, and has noted that customer meters are not required to be monitored. Please
see Section II.F of the preamble to todays final rule for further details.

Onshore petroleum and natural gas production CAPP believes the all inclusive nature of this
definition adds confusion to the already complex and burdensome reporting requirements for the
oil and gas sector and may introduce overlap with other facility definitions. Therefore, CAPP
recommends EPA reframe the onshore petroleum and natural gas production definition
ensuring no overlap occurs with other facility definitions and covers only key sources of interest
to the EPA.
Onshore natural gas plant CAPP believes there may be some overlap with the definition above
and that the broad scope of the definition includes sources that are not typically rolled into the
emissions at a gas plant as it includes the field gathering systems/out of plant fence line facilities.
CAPP recommends modifying the definition to include only the sources that are normally
associated with the physical gas plant then incorporate the other associated emissions into
another source or into a pre-existing source such as onshore natural gas transmission
compression.
Onshore natural gas transmission compression As this definition includes the equipment other
than that typically associated with compression, CAPP recommends the EPA add the word
system to the end of the sources: Onshore natural gas transmission compression system.
Response: In todays final rule, EPA has clarified the boundary between onshore production and
other segments in Subpart W. For further details, please see response to EPA-HQ-OAR-2009
0923-1004-14.
With regards to the facility definition for gas processing, in todays final rule EPA does not
include gathering lines and boosting stations Please see Section II.F of the preamble to todays
final rule and Topic 2: Aggregation of Gathering and Boosting Systems with Processing
Facilities in Volume 9 of the Response to Comments for further details. The commenter does not
provide sufficient details as to which sources not associated with a physical facility have been
included in the rule or how they can be incorporated into transmission compression stations, as
also why the word system should be added of the source category name for transmission
segment, hence EPA does not have a response. However, the definitions for natural gas
1621
processing facilities and transmission compression stations have been developed considering the
operational boundaries of the two segments and hence, in todays final rule EPA has retained
these definitions.

92.238 Definitions: Offshore petroleum and natural gas production facility means each platform
structure and all associated equipment as defined in paragraph (a)(1) of this section. All
production equipment that is connected via causeways or walkways is (sic) one facility.
Paragraph (a)(1) is assumed to be 98.230(a)(1), which states, offshore production includes
secondary platform structures and storage tanks associated with the platform structure.
OOC Comment: The dual definitions in the rule need to be aligned with each other, and the
definition in 92.230(a)(1) should also reference connection via causeways or walkways in its
source definition. Does this mean the EPA wants us to include secondary platform structures
with the facility (reporting entity) whether or not they are connected via causeways or
walkways? Further clarity is needed on the definition of offshore petroleum and natural gas
production facility considering these multiple definitions.
Response: In todays final rule EPA has consolidated the definitions. Please see Section 98.230
Definition of the source category of todays final rule for further details. Also, EPA clarifies
that any secondary structures or equipment connected to the main platform via walkways are
considered to be a part of the main platform for the purposes of threshold determination and
emissions reporting.

The key issues in this proposed rulemaking revolve around getting the definitions right. EPA
must define what constitutes the definition of a facility in the context of the nations various
petroleum and natural gas sectors.
Response: EPA does not agree with the commenter that facilities in the petroleum and natural
gas sectors are not well defined. Please see Section 98.6 of The Final Mandatory GHG Reporting
Rule (Final MRR), (40 CFR part 98), for all segments of the petroleum and natural gas
industry, except onshore production and LDCs that are defined in Sections 98.230 and 98.238 of
todays final rule.
1622

EPAs Proposed Definition of Facilities is Functionally Inappropriate and
Counter to CAA
equipment associated with all petroleum or natural gas production wells under common
petroleum and natural gas production facility. 243
We are troubled by EPAs expansive definition of facilities, when it comes to onshore
petroleum and natural gas production. It appears that EPA is attempting to treat oil and gas
production in the same way it is approaching electric power systems under another of its
pending supplemental rules to the MRR. 244
There, EPA is suggesting that all owners and operators of equipment in an electrical system be
treated as a facility, for purposes of determining whether the threshold of 25,000 tons of
emissions/year for hexafluoride and perfluorocarbons is met and reporting is required. Here, the
agency seeks to apply the same approach to natural gas production facilities, arguing for basin,
or at least field-level, reporting of GHG emissions. The problem: the two examples are an apple
to orange comparison.
The electrical equipment being discussed under the other rule is all part of an integrated
electrical system. Here, the situation is very different. The mere presence of onshore production
facilities in the same basin is not enough to link facilities. Those wells are not part of an
integrated system. Each well operates independently. They are more analogous to power
generation stations. Such facilities may sit in the same region of the country. Specific entities
may have ownership shares in more than one. Yet, the CAA treats each plant as an individual
facility. In this way, oil and gas production facilities fit much more neatly into the CAA
existing definitional structure.
Yet, the CAA itself contains a caveat when it comes to regulation of oil and gas facilities.
Section 112 lays out definitions of stationary sources whose emissions are to be regulated by
the Act. Important in this regard is the following language:
243
75 Fed. Reg. 18608 at page 18614.
244
75 Fed. Reg. 18652, April 12, 2010.

1623
emissions from any oil and gas exploration or production well (with its associated
equipment) and emissions from any pipeline compressor or pump station shall not be aggregated
with emissions from other similar units, whether or not such units are in a contiguous area or
under common control, to determine whether such units or stations are major sources, and in the
case of any oil and gas exploration or production well (with its associated equipment), such
emissions shall not be aggregated for any purpose under this Section. 245
Response: Please see Topic 1: GHG Reporting under Subpart W and the Consolidated
Appropriations Act in Volume 9 of the Response to Comments and responses to EPA-HQ-OAR
2009-0923-1044-1 and EPA-HQ-OAR-2009-0923-1005-2.

EPA should not adopt a Subpart W definition of facility that is inconsistent with the
facility definition used in other Clean Air Act programs. Industry and regulators
currently have a very good understanding of what the term facility means in the upstream
and midstream sectors. Company representatives as well as state regulatory agencies have
been using this term to apply current Clean Air Act (CAA) requirements across the
country. EPA should not introduce a new definition for purposes of this subpart. For the
upstream sector, EPA should delete the reference to a facility and should instead propose
language to refer to a threshold reporting area or roll-up area threshold to define a
geographical area to which a threshold can be applied that is made up of oil and gas upstream
facilities. By doing so, EPA would minimize the confusion and regulatory uncertainty that
might ensue should EPA finalize the rule as proposed.
For example, proposed section 98.238 Onshore petroleum and natural gas production
facility would be revised to read: Onshore petroleum and natural gas production reporting
area means all petroleum or natural gas facilities and equipment associated with all
petroleum or natural gas production under common ownership or common control by an
onshore petroleum and natural gas production owner or operator located in a single
hydrocarbon basin as defined.
98.230(2) Onshore petroleum and natural gas production: Onshore petroleum and natural
gas production reporting area means all facilities associated with the production of petroleum
or natural gas including but not limited to.
245
The Clean Air Act, Section 112(n)4)(A).

1624
Response: EPA does not agree with the commenter. For further details, please see Topic 1:
GHG Reporting under Subpart W and the Appropriations Act in Volume 9 of the Response to
Comments and response to EPA-HQ-OAR-2009-0923-1044-1.
EPA does not agree with the use of the term reporting area instead of facility. For further
details, please see the response to EMAIL-0001-1 (comment also located in rulemaking memo
Early Comment Submissions in docket EPA-HQ-OAR-2009-0923).
98.230(a)(2): The Onshore petroleum and natural gas production definition includes well
drilling, completion, and workover equipment. As explained in section 3 above, reporting should
not be required for well drilling, completion. and workover. Similarly, as explained in Section 4,
in cases where the well site operator contracts these tasks to a third-party company, the well site
operator does not own or operate the well drilling, completion, and workover equipment. When
the well site operator does not own the equipment or control the operation or maintenance of the
equipment, it is not appropriate to require reporting and compliance tasks of the well site
operator because they do not participate in engine maintenance or the collection of fuel use data.
All other CAA programs establish applicability based on whether a party owns and operates a
source because it is not feasible for a person lacking control of a source's day-to-day operation to
collect the required information or monitor the source's usage.
Response: EPA does not agree with the commenter. For further details, please see the response
to EPA-HQ-OAR-2009-0923-1031-21. EPA does not agree that reporters should not be required
to report emissions from drilling, completion, and workover operations. See response to EPA

EPA should not adopt a Subpart W definition of facility that is inconsistent with the facility
definition used in other Clean Air Act programs.
Industry and regulators currently have a very good understanding of what the term "facility"
means in the oil and gas upstream sector. Company representatives as well as state regulatory
agencies have been using this tern to apply current CAA requirements across the country. Our
recommendation is that we rely on the current actively used definition of facility and that EPA
change their proposed language to refer to a "threshold reporting area" or "roll-up area threshold"
to define a geographical area to which a threshold can be applied that is made up of oil and gas
upstream facilities. This would help alleviate industry concerns that the reporting rule definition
of "facility" might impact perception of the CAA definition.
1625
For example, 98.238 'Onshore petroleum and natural gas production facility' would read:
'Onshore petroleum and natural gas production reporting area means all petroleum or natural gas
facilities and equipment associated with all petroleum or natural gas production under common
operator located in a single hydrocarbon basin as defined....'
98.230(2) 'Onshore petroleum and natural gas production': Onshore petroleum and natural gas
production reporting area means all facilities associated with the production of petroleum or
natural gas including but not limited to ....'
Response: EPA does not agree with the commenters suggestion about facility definitions. For
further details, please see Topic 1: GHG Reporting under Subpart W and the Consolidated
Appropriations Act in Volume 9 of the Response to Comments and response to EPA-HQ-OAR
2009-0923-1044-1.
EPA does not agree with the use of the term reporting area or other such terms instead of
facility. For further details, please see the response to EMAIL-0001-1 (comment also located
in rulemaking memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923).

Organization: Spectra Energy Corp
Commenter: Brianne Metzger-Doran
The Definition of Onshore Natural Gas Processing Plant is Overbroad to the Extent it Includes
Booster Stations Where No Natural Gas Processing is Actually Performed
The proposed definition of Onshore Natural Gas Processing Plants in 98.230(3) is overly
broad and confusing to the extent the definition includes booster stations where no natural gas
processing actually occurs. Spectra Energy operates a number of booster stations that receive
pipeline quality natural gas (that is, gas that requires no processing) and pump it directly into
transmission pipelines. These facilities are distinct from processing plants designed for
separation and recovery, and from field gathering and boosting stations that feed gas processing
plants. The second to last sentence of 98.230(3) states:
Gathering and boosting stations that send the natural gas to an onshore natural gas transmission
compression facility, or natural gas distribution facility, or to an end user are considered stand
alone natural gas processing facilities.
Gathering and booster stations that receive and move pipeline quality gas into transmission or
distribution, where no additional processing occurs, are functionally equivalent to natural gas
transmission compression stations, and should be categorized accordingly for purposes of
Subpart W. Spectra Energy strongly recommends that the EPA eliminate the above quoted
sentence from 98.230(3), and/or insert language in 98.230(4) clarifying that a booster station
1626
that receives and moves pipeline quality natural gas without performing any natural gas
processing is classified as onshore natural gas transmission compression.
Response: Todays final rule does not include gathering lines and boosting stations. For further
clarification, please see Section II.F of the preamble to todays final rule.

Gas Well (EPA does not define)
Add, Gas Well means a well completed for production of natural gas from one or more gas zones
or reservoirs. Such wells contain no completions for the production of crude oil.
EPA proposed definition: Unconventional Well means a gas well in producing fields that
employ hydraulic fracturing to enhance gas production volumes.
A time element is not associated with this definition and this makes it subject to misapplication.
For example, if a gas well was fractured one or two years ago but is worked without being
fractured in a subsequent year, it could be placed into the unconventional well category for the
reporting year though that does not seem to be EPAs purpose. We request that EPA modify
the definition to:
Unconventional Well means a gas well in producing fields that is undergoing hydraulic
fracturing to enhance gas production volumes.
Response: EPA agrees with the commenter. In todays final rule, EPA has provided a definition
for natural gas wells. Also, with regards to well completions and workovers, EPA has revised the
source category names that now distinguish the source based on whether or not the completion
involved hydraulic fracturing or not in the reporting year. This will avoid any confusion relating
to the timing of when the hydraulic fracturing was conducted.

In regards to the proposal to require reporting of greenhouse gas (GHG) emissions including
1627
petroleum and natural gas systems, COGAs main concern is the restrictive and inconsistent
definition of facility as previous defined in previous rules and established in the Clean Air Act.
2009-0923-1044-1 and EPA-HQ-OAR-2009-0923-1005-2.

Usage of facility for natural gas distribution, offshore petroleum and natural gas production
and onshore petroleum and natural gas production.
Incorporating the term facility in this definition adds confusion to the term currently well
understood by industry and regulating bodies. The Clean Air Act as well as 98.6 of this Part
define a facility as any physical property, plant, building, structure, source, or stationary
equipment located on one or more contiguous or adjacent properties Our recommendation is
to replace facility with reporting unit or area which would encompass the applicable
facilities encompassed in the producing basin.
Response: EPA does not agree with the commenters suggestion about facility definitions. For
further details, please see Topic 1: GHG Reporting under Subpart W and the Consolidated
2009-0923-1044-1.

Organization: Vorys, Stater Seymour and Pease LLP
Commenter: Gregory D. Russell
While the boundaries of overall production operations can vary widely and may not be easily
characterized, if greenhouse gas emissions monitoring and reporting are going to be required of
producers - which the Association continues to oppose, particularly for small producers - the
most sensible facility definition is the one proposed by IPAA, i.e., defining the well pad as the
production facility. This tracks the nature of the producer's operations and is consistent with the
definition applied to other industries under the CAA, in addition to being the typical permitting
unit under CAA regulations.
1628
Response: EPA does not agree with the commenter. Please see responses to EPA-HQ-OAR
2009-0923-1005-6 and EPA-HQ-OAR-2009-0923-1005-2 for further details.

Comment 2: The EPA Subpart W definition of facility (i.e. basin) is inconsistent with the wellestablished facility definition currently used in other Clean Air Act programs. This definition
causes confusion and results in a much larger reporting burden for industry than anticipated by
EPA in its impact analysis.
2009-0923-1044-1 and EPA-HQ-OAR-2009-0923-1005-2.

Alternatively, EPA Must Clarify That Emissions Should Be Reported by the Operator of a Given
Well
Under the final GHG MRR, the designated representative of a facility 246 is responsible for
certifying, signing, and submitting GHG emissions reports. Significantly, the proposed
definitions of facility for onshore and offshore petroleum and natural gas production, and for
natural gas distribution, differ from the definition of facility applied in the remainder of the
Mandatory Reporting Rule. This distinction is particularly key for the onshore petroleum and
natural gas production sector. In a single geologic basin, there may be thousands of individual
sources, owned and operated by a variety of entities and contractors. Furthermore, this segment
is typified by complex operational, infrastructure, commercial and ownership arrangements, and
the fact that many of the operators are small companies without significant resources,
knowledge, and staff, making this distinction particularly critical. Likely due to these issues,
EPA has expanded the definition of an onshore petroleum and natural gas production facility
to include all equipment associated with production and distribution under common ownership or
control and located in a single hydrocarbon basin:
246
A facility means any physical property, plant, building, structure, source, or stationary equipment located on one
or more contiguous or adjacent properties in actual physical contact or separated solely by a public roadway or
other public right-of-way and under common ownership or common control, that emits or may emit any
greenhouse gas. 40 C.F.R. Section 98.6.
1629
all petroleum or natural gas equipment associated with all petroleum or natural gas production
wells under common ownership or common control by an onshore petroleum and natural gas
production owner or operator located in a single hydrocarbon basin . . . Where an operating
entity holds more than one permit in a basin, then all onshore petroleum and natural gas
production equipment relating to all permits in their name in the basin is one onshore petroleum
and natural gas production facility.
75 Fed. Reg. at 18,647. Under this definition of facility, the reporting entity for an onshore
petroleum and natural gas production facility is
the operating entity listed on the state well drilling permit, or a state operating permit for wells
where no drilling permit is issued by the state, who operates onshore petroleum and natural gas
production wells and controls by means of ownership (including leased and rented) and
operation (including contracted) stationary and portable (as defined in this Subpart) equipment
located on all well pads within a single hydrocarbon basin. Id. at 18,614.
API suggests that EPA clarify the obligation of operating entities for onshore petroleum and
natural gas production facilities to report emissions from contractors operating within a geologic
basin. Specifically, EPA should sharpen the distinction between contractors that are acting under
the direction of an operating entitywho generate emissions that arguably must be reported by
the operating entityversus contractors that are not under the operating entitys control. EPA
should make clear that the operating entity is not responsible for reporting contractor emissions
in situations where it does not operate a given well location and therefore does not necessarily
have reasonable access to emissions data.
Response: Please see the responses to EPA-HQ-OAR-2009-0923-1031-21 and EPA-HQ-OAR
2009-0923-1170-7.

Onshore Petroleum and Natural Gas Production Facility Definition
Although API believes that grouping owner/operators GHG emissions at the American
Association of Petroleum Geologists (AAPG) basin level may be useful for threshold
determination and reporting purposes under this rule, we do not believe it is beneficial from the
standpoint of balancing burden with quality emissions data. EPAs focus on millions of
individual sites and equipment is adding complexity to the effort of compliance with the rule
without necessarily increasing the data quality. API believes the issues can be resolved by
allowing owners/operators to group together similar operations and sources for emissions
calculation rather than forcing emissions estimation on every source within a basin.
Most basins are complex and can have several different types of production, multiple dissimilar
1630
reservoirs, with broadly differing fluid compositions and make-up, and different equipment and
operational practices driven by the reservoir characteristics and type of production. This
complexity poses a real challenge to developing emissions estimation methods which
appropriately balance burden and quality. Allowing the grouping of fields (in EIAs field list)
within a basin based on common production characteristics will achieve EPAs articulated goal
in the proposed rule of reducing burden without compromising data quality. These characteristics
would be the same (or groups of similar) producing reservoirs, similar well depths and well-bore
configurations, similar pressure and temperature ranges, similar fluid compositions and Gas-toOil Ratios (GORs), similar production arrangements and surface equipment, and similar
operational characteristics and practices. For example, in the state of Wyoming a similar site is
defined for hydrocarbon sampling based on the same reservoir and production equipment with
similar characteristics (25 psig), allowing compositional analysis to be determined from a
minimum of 5 wells and an updated analyses conducted every 3 years.
Each covered operator in a basin would select these groupings of fields based on their particular
operations and portfolio, using the above characteristics, and fully describe them in their
monitoring plan along with their rational for a particular grouping. This Sub-basin entity
would enable the development of activity or emission factors and approaches based on the
parameters that affect the amount and type of emissions from a particular Sub-basin area.
Although the proposed rule does enable this approach in limited instances and source types, the
approach still needs to be broadly expanded.
In our comments regarding each source type, API will detail how these activity/emission factors
and approaches may be developed for a particular source type within a Sub-basin entity grouping
of similar fields. Examples of how these Sub-basin entities might be demonstrated, could be
discussed further with EPA.
i. Facility Terminology
API concurs with EPAs statement in the FAQs to proposed Subpart W (Subpart W FAQ, March
2010) that the facility definitions proposed in this rule do not impact requirements under other
EPA regulations, for example, New Source Review (NSR). The structure of the Clean Air Act
(CAA) supports EPAs statement and in fact requires this result. EPA relies on its authority
under Section 114 of the CAA as the basis for the proposed GHG reporting ruleauthority
which is completely independent from EPAs authority for other CAA programs, for example,
from EPAs CAA NSR authority, including CAA Sections 160-169 and 171-193. As such, EPA
regulations promulgated under these distinct provisions in Section 114 cannot and should not
impact the other CAA programs and vice versa. Therefore, EPAs definition of facility for
purposes of the proposed Subpart W in no way impacts the facility definition for similar
sources under the NSR program or other existing CAA programs.
While we appreciate that EPA has clarified that the proposed definition of facility in this rule
does not impact other EPA regulations, we do not feel that the FAQ response provides enough of
a backstop against a slow creep in the definition under other CAA regulatory programs. EPAs
statement should be included in the regulation itself.
1631
Furthermore, API suggests an alterative term that is not already defined elsewhere in the CAA be
proposed (in place of facility) for the geological basin grouping of emissions. We propose the
term Basin entity be used in place of facility for all discussions related to the geologic basin
grouping of emissions that are made up of distributed sources either at the field level or for
sources that are grouped across production fields. We propose the term Sub-basin entity for
describing the grouping of fields with similar characteristics within the basin. API also
recommends that the term facility only be used when referring to a traditional facility
(consistent with CAA terminology), such as an oil and gas processing plant, to avoid confusion.
API Requests:
- EPA should enable the grouping of similar named fields within a basin into a Sub-basin
entity for emissions determination based on the characteristics outlined above.
- EPA should use the term Basin entity when referring to the roll-up of Sub-basin entities for
assessing onshore oil and gas production operations against the reporting threshold.
- EPA should add the statement that the facility definitions in this rule do not impact
requirements under other EPA regulations, for example, New Source Review (NSR) to the
regulation itself.
Response: EPA disagrees with the commenter on the grouping of sub-basin entities. For further
clarification, please see the response EPA-HQ-OAR-2009-0923-1305-46.
As regards impact of this rule on other EPA regulations, please see response to EPA-HQ-OAR

Portable Equipment Consistent with WCIs comments to EPA, we also support the inclusion of
Portable Equipment in the Subpart W petroleum and natural gas production definition. We agree
with WCI that Portable Equipment should be defined as all portable non-self propelled
equipment used in the petroleum and natural gas production industry. We agree the definition
should not be limited to equipment stationed at a wellhead for more than 30 days. Portable
Equipment (e.g. drilling rigs, workover rigs, construction equipment) can generate substantial
emissions in periods of less than 30 days. Subpart W should not create an incentive for Operators
to stage Portable Equipment at a location for less than 30 days, to thereby avoid Portable
Equipment GHG emissions reporting. If this occurred, it would severely underestimate the actual
emissions being generated.
Response: EPA agrees with the commenter. In todays final rule, EPA has removed the 30 day
at location requirement for portable equipment emissions monitoring. Please see the response to
1632

The concept of Reporting Area should replace the proposed definition of onshore petroleum
and natural gas production facility. The Subpart W definition of facility for onshore petroleum
and natural gas production is inconsistent with the facility definition for other CAA programs
and will cause undo complexity for companies managing compliance with numerous air
regulations. Designation of reporting areas for this Subpart W will establish the source
category that EPA desires without using an overly board definition of facility that will result in
confusion, uncertainty and practical application problems. The Section 98.6 definition of facility
should be retained and apply to onshore petroleum and natural gas production.
The proposed term onshore petroleum and natural gas production facility (proposed 40 C.F.R.
Section 98.238) as currently drafted is overly-broad and will create legal and practical
application uncertainty. First, the term in section 98.238 is overtly inconsistent with the
definition of facility in 98.6, and the definition of facility as used under other Clean Air Act
regulatory programs. The EPAs statement on its web posting (Subpart W FAQ, March 2010)
states that . . . the facility definitions proposed in this rule do no impact requirements under
other EPA regulation, for example, New Source Review (NSR). Noble appreciates that the EPA
will limit this novel facility definition to reporting under Subpart W. However, this deviation
from historical regulatory precedent and delineation of source boundaries and revised application
of the term facility for this one GHG regulatory scheme will create undue confusion not only
by the regulated community, but also among regulators. It further subjects both the regulated
community and the EPA to litigation risk regarding the meaning and use of this term. Second,
Subpart W appears to be the only subpart that imposes an expansive definition of facility. All
other subparts related to other source categories adhere to the term facility as defined in 40
C.F.R. Section 98.6.
The objective of the proposed rule is to establish reporting requirements for facilities that emit
greenhouse gases as contemplated by 40 C.F.R. Section 98.1(a). This objective can be achieved
for source categories subject to Subpart W without complicating the use of the term facility. To
this end, Noble recommends the term onshore petroleum and natural gas production facility in
Section 98.238 be deleted in its entirety and replaced with a reporting area. This concept of a
reporting area is already being utilized by EPA in its currently proposed definition of onshore
petroleum and natural gas production facility, however it is used in a way that does not
distinguish it from a facility. More clarity and regulatory certainty is provided by clearly
distinguishing reporting area from facility in this regulatory context. . Therefore, Noble
recommends that the definition of onshore petroleum and natural gas production facility in
Section 98.238 be amended and replaced with the following proposed definition of onshore
petroleum and natural gas production reporting area:
Onshore petroleum and natural gas production facility reporting area means all facilities that
1633
contain petroleum or natural gas equipment associated with all onshore petroleum or and natural
gas production wells under common ownership or common control by an onshore petroleum and
natural gas production owner or operator located in a single hydrocarbon basin as defined by the
American Association of Petroleum Geologists which is assigned a three digit Geologic Province
Code. Where an operating entity onshore petroleum and natural gas production owner or
operator holds more than one permit in a basin, then all onshore petroleum and natural gas
production equipment relating to all permits in their name in the basin is one included in the
same onshore petroleum and natural gas production facility reporting area.
This proposed definition of reporting area clearly defines the geographic reporting boundaries
for an owner or operator of onshore petroleum and natural gas production sources which are
subject to the reporting requirements without creating confusion and uncertainty regarding the
term facility or unintentionally expanding the meaning of facility.
Response: EPA does not agree with the commenter on the inconsistency of facility definition not
being compatible with the CAA. For further details, please see Topic 1: GHG Reporting under
Subpart W and the Consolidated Appropriations Act in Volume 9 of the Response to Comments
and response to EPA-HQ-OAR-2009-0923-1044-1.
Early Comment Submissions in docket EPA-HQ-OAR-2009-0923). As regards impact of this
rule on other EPA regulations, please see response to EPA-HQ-OAR-2009-0923-1174-5 for
further details.

Section 98.2 Who must report?
An owner or operator of onshore petroleum and natural gas production facilities that have total
emissions from all facilities located within an onshore petroleum and natural gas production
reporting area (as defined in Section 98.238) of 25,000 metric tons CO2e or more per year.
If EPA does not accept Nobles recommended revisions to the Proposed Rules definition of
offshore petroleum and natural gas production facility, Noble believes it is imperative that EPA
include language in the Rule to state that the Rules definition of offshore petroleum and natural
gas production facility will not be applied elsewhere in the CAA and will not impact other EPA
regulations.
Note to reader, in the comments that follow the familiar facility is often retained for
consistency with the Proposed Rule language. As discussed above, facility would be replaced
by reporting area in the final rule.
1634
Response: EPA does not agree with the commenter with regards to restrictions on the use of
definition from Subpart W in other EPA regulations; please see response to EPA-HQ-OAR

Onshore petroleum and natural gas production owner or operator means the entity who is the
permitee to operate petroleum and natural gas wells on the state drilling permit or a state
operating permit where no drilling permit is issued by the state, which operates an onshore
petroleum and/or natural gas production facility (as described in Section 98.230(b)(2) Section
98.230(a)(2)). Where more than one entity are permitees on the state drilling permit, or operating
permit where no drilling permit is issued by the state, the permitted entities for the joint facility
must designate one entity to report all emissions from the joint facility holds legal or equitable
title to or control over a source in section 98.230(a)(2), the entity exercising the greatest
operational control over the source category shall be deemed the owner or operator designed to
report all emissions. For all purposes of Subpart W, onshore petroleum and natural gas
production owner or operator excludes any other person or entity who has legal or equitable title
to, has a leasehold interest in, or control of a facility or supplier or petroleum and natural gas
well or source, including but not limited to royalty interest owners and non-operators.
Noble recommends that the language above, or similar text, be incorporated into the definition
for onshore petroleum and natural gas production owner or operator in Section 98.6.
Response: EPA does not agree with the commenter. EPA will not constrain owners or operators
in choosing who shall be the designated representative. The designated representative (DR) is the
entity that is responsible for submitting the emissions data pursuant to todays final rule. Please
see the responses to EPA-HQ-OAR-2009-0923-1024-16 and EPA-HQ-OAR-2009-0923-1031
21.

The definition of onshore petroleum and natural gas production owner or operator should be
revised to clarify that royalty owners and non-operating interest owners be excluded from the
revised definition.
1635
Response: The issue highlighted by the commenter has already been dealt with in the final
MRR. Please see the preamble Section V.B.1 response on the Certification Statement of the
Final MRR (October 2009), and rule Sections 98.4(b) and 98.4(i)(4)(iv) of the Final MRR.

Section 98.232 lists eight Subpart W industry segments [i.e. (b) through (i)] and the emission
sources to report for each segment. For onshore petroleum and natural gas production, 21
primary sources are identified. Noble supports EPAs intent to limit reporting to segmentspecific sources listed in Section 98.232. However, additional clarifying rule text is needed to
avoid unnecessary implementation questions.
Onshore petroleum and natural gas production and processing operations include an array of
processes and equipment / source types, ownership, leasing, and arrangements that significantly
complicate clear facility and source applicability delineation. For example, a central gathering
facility that has liquid stabilization may have sources from onshore petroleum and natural gas
production Section 98.232 (c), and also onshore natural gas processing Section 98.232 (d). In this
example, the lack of clear segment delineation from the multi-use facility unnecessarily
complicates rule interpretation, implementation and compliance. Noble understands that a
facility is only required to report emissions from the sources listed in the applicable Section
98.232 subsection; that is, the source list is specifically defined and limited to those sources in
the Section 98.232 subsection for that segment.
Noble further understands that the applicable segment is based on the primary facility function.
The 21 emission sources in Section 98.232(c) are to be reported under Subpart W for onshore
petroleum and natural gas production. If other emission sources applicable to another segment
(i.e. beyond the 21 onshore production sources) are at a production facility (e.g., a blowdown
vent stack), emissions reporting for that source is not required.
Further clarification directed toward accurately defined segment definition and source
applicability is needed for unambiguous identification of over riding section of the rule. Noble
does not advocate reporting under multiple segments and instead advocates a primary facility
reporting segment and source requirements. These clarifications are required to ensure rule
implementation and compliance issues do not arise.
Response: In todays final rule EPA has provided clear requirements on reporting of emissions
from co-located facilities and multiple use equipment. For further details, please see the response
to EPA-HQ-OAR-2009-0923-1024-14. EPA has also reviewed and clarified as necessary all
source category definitions and concluded that todays final rule should mitigate any concerns
regarding ambiguity in definitions. Todays final rule defines the processing industry segment as
the processing facility and does not include gathering lines and boosting stations. For further
clarification, please see Section II.F of the preamble to todays final rule. Hence the overlap issue
1636
between onshore production and onshore natural gas processing does not apply anymore. For
further clarification on gathering and boosting systems, please see Section II.F of the preamble to
todays final rule.

The definition of facility" in the proposed Subpart W, upon which an operator would assess
applicability of the 25,000 tons/year greenhouse gas ("GHG") threshold, includes all petroleum
and natural gas equipment associated with all production wells under common ownership or
common control by an onshore petroleum and natural gas production owner or operator located
in a single hydrocarbon basin. This definition is inconsistent with the definition of facility" in
other subparts of this rule, and therefore will result in an inequitable application of the rule. Also,
it is inconsistent with the "facility" definition in the Clean Air Act ("CAA"). The preamble states
that the information gathered by the implementation of this rule would inform and guide EPA in
carrying out a wide variety of CAA provisions. It is understood that EPA will use the GHG
inventory data to define future permitting and compliance programs. For this reason, we do not
agree with redefining the term "facility" in this Subpart. If EPA uses the information in this
program for other CAA programs, it must be collected in a manner that is consistent with how it
will be applied in those programs.
GHG Reporting under Subpart W and the Consolidated Appropriations Act in Volume 9 of the
Response to Comments and responses to EPA-HQ-OAR-2009-0923-1044-1 and EPA-HQ-OAR
2009-0923-1005-2.

Building, structure, facility or installation means all of the pollutant-emitting activities which
belong to the same industrial grouping, are located on one or more contiguous or adjacent
properties, and are under the control of the same person (or persons under common control)
except the activities of any vessel. Pollutant-emitting activities shall be considered as part of the
same industrial grouping if they belong to the same "Major Group" (ie: which have the same first
two digit code) as described in the Standard Industrial Classification Manual, 1972, as amended
by the 1977 Supplement (U.S. Government Printing Office stock numbers 4101-0066 and 003
005-00176-0, respectively).
In accordance with this definition and multiple case-by-case aggregation determinations made by
1637
the EPA and the states, most production operations across a single hydrocarbon basin cannot be
aggregated into one facility. If these production operations are not aggregated by basin, each
facility or well site must be individually assessed for applicability to the future permitting and
compliance programs. At this time, EPA has proposed an applicability threshold of 25,000
metric tons CO2e/year for permitting and that threshold will apply at the facility level, as
"facility" is currently defined in the Clean Air Act. If the purpose of this rule is to inform future
rule-making, there is no benefit to collecting information for an aggregation of many small
facilities that will not be subject to future rules. We therefore request that EPA modify this
"facility" definition to be consistent with the "facility" definition employed in other CAA
programs.
Based on this basin-wide "facility" definition, we anticipate that essentially all of Pioneer's asset
areas in the U.S. would be subject to reporting under Subpart W. This basin-wide application
would aggregate many insignificant emissions sources together and is not reasonable from a
cost/benefit analysis of supplying representative, useful, high quality emissions data. The time
and cost burden of just collecting this data would be massive, as outlined in the points below. In
order to achieve a lower burden and more representative, high quality emissions data, Pioneer
agrees with API's recommended approach that would allow operators the flexibility to report at
the basin level while also allowing them to identify rational "reporting units" within a basin (that
have similar operations, equipment, fluid properties, and physical properties).
2009-0923-1005-2.
.
EPA disagrees with the commenter on the grouping of sub-basin entities. For further
clarification, please see the response EPA-HQ-OAR-2009-0923-1305-46.
Definition of "Facility" for Onshore Petroleum and Natural Gas Production
"EPA seeks comments on our decision to propose the basin level approach and whether there
would be advantages to requiring reporting at the field level instead. " (75 FR 18615)
Resolute Comments:
Resolute opposes EPA's proposed definition of "Onshore petroleum and natural gas production
facility," as it is inconsistent with other regulatory programs under the CAA. Industry and
regulators currently have a very good understanding of what the term "facility" means in the oil
and natural gas upstream sector, as they have been using this term since it was promulgated in
1638
1980 with the Prevention of Significant Deterioration ("PSD") regulations. Resolute

recommends that EPA use this well-established definition of facility instead of introducing an
entirely new definition in Proposed Subpart W. EPA should alter the proposed language in
Proposed Subpart W to maintain the traditional definition of "facility" but then refer to a
"threshold reporting area" or something similar to define a geographical area (e.g., basin-wide or
field-wide). An owner/operator would then need to calculate the greenhouse gas emissions from
each of its onshore petroleum and natural gas production facilities within the threshold reporting
area to determine if it exceeds the reporting threshold and is subject to the reporting
requirements. This would help alleviate industry concerns about how the Proposed Subpart W
definition of "facility" might impact this same term as defined in other CAA programs, in
particular New Source Review.
As an example, Resolute would suggest that EPA make the following changes to Proposed
Subpart W 98.238 and 98.230(2):
98.238: Onshore petroleum and natural gas production threshold reporting area means all
petroleum or natural gas facilities and equipment associated with all petroleum or natural gas
production under common ownership or common control by an onshore petroleum and natural
gas production owner or operator located in a single hydrocarbon basin as defined ....
98.230(2): Onshore petroleum and natural gas production: Onshore petroleum and natural gas
production equipment threshold reporting area means all structures facilities associated with the
production of petroleum or natural gas including but not limited to all structures associated with
wells ....
In the alternative, if EP A decides to proceed with the definition of "Onshore petroleum and
natural gas production facility" as proposed, Resolute requests that EPA expressly state that "the
facility definitions proposed in this rule do not impact requirements under other EPA regulations,
for example New Source Review (NSR)" in the rule itself, as it did in the FAQ for Proposed
Subpart W, or at a minimum provide a similar statement in the preamble to the final rule when it
is published in the Federal Register.
Response to Comments and response to EPA-HQ-OAR-2009-0923-1044-1.
EPA does not agree with the use of the term threshold reporting area instead of facility. For
memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923). As regards impact
of this rule on other EPA regulations, please see response to EPA-HQ-OAR-2009-0923-1174-5
for further details.
1639

EP A should use a definition of "facility" for onshore oil and natural gas production that is
consistent with other CAA regulatory programs, but provide for basin-level reporting by rolling
up facility emissions into a basin-wide "reporting area."
memo Early Comment Submissions in docket EPA-HQ-OAR-2009-0923) and response to
EPA-HQ-OAR-2009-0923-1044-1.

Devon does not support the proposed onshore petroleum and natural gas production facility
definition for the following reasons.
First, aggregation of upstream production sites imposes a disproportionate burden on our
industry by requiring data collection, recordkeeping, and reporting for hundreds of thousands of
dispersed production sites.
Second, EPAs proposal to aggregate all E&P wells within a basin as a single onshore
petroleum and natural gas production facility is inconsistent with existing Clean Air Act (CAA)
regulations and incongruous with ownership and control of affected operations.
Third, we are concerned that EPAs definition of onshore petroleum and natural gas production
facility could potentially impact the facility definition for similar sources under other CAA
regulatory programs. Although we agree with EPAs statement in the Frequently Asked
Questions to proposed Subpart W that the facility definitions proposed in this rule do not impact
requirements under other EPA regulations, for example, New Source Review (NSR), we
believe it is imperative that this statement is included in the regulation itself.
Response: Without more than a broad comment that the basin-level definition of facility with
respect to onshore petroleum or natural gas production imposes a disproportionate burden, it is
difficult for EPA to respond in detail. Please see Volume 10 of responses to comment and EIA
to todays final rule for details on how EPA has determined that the burden to report to the rule is
reasonable.
EPA does not agree with the commenter on inconsistencies with the CAA. For further details,
please see Topic 1: GHG Reporting under Subpart W and the Consolidated Appropriations Act
in Volume 9 of the Response to Comments and response to EPA-HQ-OAR-2009-0923-1044-1.
Likewise, the broad statement that the proposal to aggregate all E&P wells within a basin is
inconsistent with the CAA and regulations without more precise explanation, makes it difficult
for EPA to respond. Generally, however, as explained above and in the preamble to the
1640
proposed rule (75 FR 18608) and accompanying Technical Support Document (TSD) found in
docket (EPA-HQ-OAR-2009-0923-0027), because of the unique nature of petroleum and natural
gas production facilities, it is reasonable and appropriate to define the facility based on the
hydrogeologic basin and equipment owned or operated by the entity holding permits in that
basin. The physical boundaries of a production or distribution owner or operators holdings are
not discrete such as for other industry segments. And while the PSD and NSR programs might
apply to stationary sources as defined by regulation, the scope of CAA Section 114 is not so
limited; nor is EPA regulating greenhouse gases under the MRR. The scope of the rule and the
authority on which is based is not limited to regulated sources and to graft a source definition
from other regulatory programs on the rule is neither appropriate nor reasonable. Further, as the
owner or operator is only required to report information from its own wells or operations in the
basin, and as the owner or operator is defined in todays final rule as the person or entity who
holds the drilling or operating permit (or otherwise who pays the state or federal business income
taxes for the operation), EPA disagrees that any incongruity with ownership or control exists.
The definition of facility with respect to onshore petroleum or natural gas production in todays
final rule expressly states it is defined, for purposes of this subpart and subpart A. Further,
Section 98.238 of todays final rule states that [e]xcept as provided below, all terms used in this
subpart have the same meaning given in the Clean Air Act and subpart A of this part. These
provisions separately stated, and particularly when read together, make it clear that the facility
definition is discrete to Subpart W. EPA considers this language to sufficiently address
limitation of the definition and its relationship to other programs.

Organization: West Virginia Oil and Natural Gas Association
Commenter: Nicholas DeMarco
As drafted, Subpart W would essentially require the reporting of GHG emissions from every
source, no matter the size, due to the definition of "facility".
Response: In todays final rule, EPA has provided equipment thresholds that require the use of
simple emissions factors for equipment below the thresholds. This will avoid unnecessary burden
on smaller sources while at the same time gather necessary data to inform policy. Please see
Section II.F of the preamble to todays final rule for further details.

Defining "Facility"
1641
One of the most concerning aspects of this proposal is the EPA's various proposed definitions of
"facility". EPA has proposed numerous definitions depending on the industry sector. Not only
does this cause confusion in determining applicability, the definitions would impose significant
burden on the industry as a whole. We agree with AXPC that EPA rely on the current definition
of "facility" which has been used by both industry and state agencies regularly. We are
concerned with the proposal to redefine the universe of facilities within the natural gas sector for
purposes of this rule and we agree with the comments submitted on behalf of GPA and AXPC.
We are particularly concerned with the proposal to aggregate emissions at the basin level and
with the proposed definition of natural gas processing facility, as set forth in more detail below.
Response: EPA does not agree with the commenter. EPA has chosen the onshore definition to
collect necessary information to inform policy. Please see response to EPA-HQ-OAR-2009
0923-1044-1 for further details. EPA has also revised onshore gas processing definition to clarify
any overlap issues with onshore production; please see response to EPA-HQ-OAR-2009-0923
1080-45 for further details. Finally, EPA has clarified the boundaries of onshore production as it
relates to other segments. For further details, please see response to EPA-HQ-OAR-2009-0923
1004-14.

As mentioned above, the SUGF does not agree, in particular, with the Subpart W definition of
facility. Primarily, the redefining of the term facility in this proposal is not consistent with the
way the term is defined and utilized in numerous Clean Air Act (CAA) Programs. As proposed,
the definition of facility will require aggregation of numerous production sites. It is for this
reason, and the perception that this may create when used in other CAA programs that the SUGF
disagrees.
If the Greenhouse Gas (GHG) emissions inventory data is to be utilized as understood (i.e. to
define future permitting and compliance programs); then data should be collected in a manner
consistent with how it will be applied, as historically completed with well established CAA
programs. Therefore, the SUGF recommends EPA utilize a different term to describe the
proposed reporting area(s), such as, basin and/or field "reporting area".
2009-0923-1044-1.
1642

Section 98.230(a)(2): The Onshore petroleum and natural gas production definition includes well
drilling, completion, and workover equipment. In cases where the well site operator contracts
these tasks to a third-party company, the well site operator does not own or operate the well
drilling, completion, and workover equipment. When the well site operator does not own the
equipment or control the operation or maintenance of the equipment, it is not appropriate to
require reporting and compliance tasks of the well site operator because they do not participate in
engine maintenance or the collection of fuel use data. All other CAA programs establish
applicability based on whether a party owns and/or operates a source because it is not feasible for
a person lacking control of a sources day-to-day operation to collect the required information or
monitor the sources usage. IPAMS requests that EPA clearly state in this part that only the
equipment a party owns and/or operates be included in that partys reported emissions inventory.
Response: EPA does not agree with the commenter. For further details, please see the response
to EPA-HQ-OAR-2009-0923-1031-21.

Section 98.238: The definition for Onshore petroleum and natural gas production facility
includes all petroleum or natural gas equipment associated with all petroleum or natural gas
production wells under common ownership or common control by an onshore petroleum and
natural gas production owner or operator located in a single hydrocarbon basin.
The preamble states that the information gathered by the implementation of this rule would
inform and guide EPA in carrying out a wide variety of CAA provisions. It is understood that
EPA will use the GHG emission inventory data to define future permitting and compliance
programs. For this reason, IPAMS does not agree with redefining the term facility in this part.
If EPA uses the information in this program for other CAA programs, it must be collected in a
manner that is consistent with how it will be applied in those programs.
Future GHG permitting and compliance programs will still rely on the definition of facility as
found in 40 CFR 52.21(b)(6):
Building, structure, facility, or installation means all of the pollutant-emitting activities which
same industrial grouping if they belong to the same Major Group (i.e., which have the same
1643
first two digit code) as described in the Standard Industrial Classification Manual, 1972, as
amended by the 1977 Supplement (U. S. Government Printing Office stock numbers 41010066
and 003005001760, respectively).
GHG Reporting under Subpart W and the Consolidated Appropriations Act in Volume 9 of

In accordance with this definition and multiple case-by-case aggregation determinations made by
EPA and the states, most production operations across a single hydrocarbon basin cannot be
aggregated into one facility. If these production operations are not aggregated by basin, each
facility or well site must individually be assessed for applicability to the future permitting and
compliance programs. EPA has proposed an applicability threshold of 25,000 metric tons CO2e
per year for permitting, and that threshold will apply at the facility level. If the purpose of this
rule is to inform future rulemaking, there is no benefit to collecting information for an
aggregation of many small facilities that will not be subject to the future rules. IPAMS requests
that EPA modify this facility definition to be consistent with the facility definition employed
in other CAA programs.
Onshore Petroleum and Natural Gas Production Facility Definition
As EPA has noted, the Onshore Petroleum and Natural Gas Production segment of the industry
is incredibly complex and diverse which makes crafting a regulation that is appropriate across
this complexity extraordinarily difficult. Additionally, it is very different than the single facility
focus of most EPA air quality regulatory focus. In the proposed Subpart W, EPAs geologic
basin approach is one of the mechanisms EPA employed to cope with these issues. BP believes
that grouping GHG emissions from a particular owner/operators activities at the geologic basin
level for wells and closely associated sites and activities has limited utility for threshold
determination and reporting purposes, we do not believe it is beneficial from the standpoint of
efficiently quantifying emissions while balancing burden with quality emissions data. EPAs
focus on millions of individual sites and pieces of equipment adds huge complexity to the effort
1644
of implementing and subsequent compliance with the rule without a corresponding increase in
the data quality. BP suggests that these issues can be partially resolved by allowing
owners/operators to group together similar operations and sources to enable population
sampling and averaging approaches to emissions calculation rather than forcing emissions
estimation on every individual source or individual piece of equipment within a basin. Most
basins are quite complex and can have several different types of production, multiple dissimilar
reservoirs, broadly differing fluid compositions and make-up, and different equipment and
operational practices driven by the reservoir characteristics and type of production. This
complexity poses a barrier to developing emissions estimation methods across a similar set of
production wells and operations which appropriately balances burden and quality. Allowing the
grouping of fields (in EIAs field list) within a basin based on common production characteristics
can enable approaches and methodologies that will achieve EPAs articulated goal in the
proposed rule of reducing burden without compromising data quality. These fields would be
grouped based on the same type of production, same (or groups of similar) producing reservoirs,
similar well depths and well-bore configurations, similar pressure and temperature ranges,
similar fluid compositions and Gas-to-Oil Ratios (GORs), similar production arrangements and
surface equipment, and similar operational characteristics and practices.
The following descriptions illustrate how these characteristics might be considered to group
fields, in a defined basin, into sub-basin reporting units.
- Production type: grouping of fields or wells within fields on the basis of gas or oil production.
- Same or similar producing reservoirs: grouping of fields on the basis of reservoir types such as
tight sands, coal bed methane, conventional sands, and shale gas. Different named
formations/reservoirs with the same classification, such as tight sands, with less than 2,000
vertical feet between the formation tops could be grouped into one reporting unit/area.
- Similar ranges of pressure and temperature for the initial phase separation of production from
the wells. Although the pressure can vary quite widely, for even the same producing
horizon/formation, dependent on well-head compression the general collection and gathering
system pressure in the fields being grouped should be similar.
- Similar fluid compositions such as oil with associated hydrocarbon gas, primary hydrocarbon
gas production with hydrocarbon liquids that separate at field separators, dry gas with no
appreciable (<2 bbls per MMSCF) hydrocarbon liquid production.
- Similar production arrangements, surface equipment, and operational characteristics/practices:
Fields to be grouped should employ similar production approaches such as well-site phase
separation with equipment located on or near individual well sites or small groups of wells,
multi-phase flow to central separation and production facilities (such as central tank batteries).
Each covered operator in a basin would select the appropriate groupings of fields based on their
particular operations and portfolio, using the above characteristics and guidance, and fully
describe them in their monitoring plan along with their rational for a particular grouping. This
sub-basin entity would enable the development of activity or emission factors and approaches
1645
based on the parameters that affect the amount and type of emissions from a particular reporting
area. Although the proposed rule does enable this approach in limited instances and source types,
the approach needs to be broadly expanded. Threshold determination and grouped reporting
would be based on the summation of the sub-basin entities within a basin.
The comments later in this document regarding specific source categories/types expand on this
concept by describing how such an approach would be applied.
Facility Terminology
To avoid the confusion which may develop regarding the definition of an Onshore Oil and Gas
Production Facility at a geologic basin level BP suggests that a different term be used rather
than facility. The use of facility should be restricted to its traditional use such as a discrete oil
and gas processing facility. BP suggests the term Basin Entity be used when referring to the
grouping of a particular owner/operators properties and activities within a basin for threshold
analysis and reporting purposes.
BP does appreciate the clarification in the frequently asked questions that the use of facility
to describe the basin level grouping of Onshore Oil and Gas Production Operations (Subpart W
FAQ, March 2010) do not impact requirements under other EPA regulations, for example, New
Source Review (NSR). However, we do not feel that the FAQ response provides adequate
clarification of this issue and request that the clarification be incorporated directly into the rule
language.
Requests:
- EPA should enable the grouping of similar named fields within a basin into a sub-basin entity
for emissions determination based on the characteristics outlined above.
- EPA should add the statement that the facility definitions in this rule do not impact
requirements under other EPA regulations, for example, New Source Review (NSR) to the
regulation itself.
- EPA should use the term Basin entity when referring to the roll-up of sub-basin entities for
assessing onshore oil and gas production operations against the reporting threshold.
Response: EPA disagrees with the commenter on the grouping of sub-basin entities. For further
clarification, please see the response EPA-HQ-OAR-2009-0923-1305-46. As regards impact of
this rule on other EPA regulations, please see response to EPA-HQ-OAR-2009-0923-1174-5 for
further details. EPA does not agree with the use of the term basin entity instead of facility.
For further details, please see the response to EMAIL-0001-1 (comment also located in
1646

Inclusion of Flare Stacks in Onshore Petroleum and Natural Gas Production
BP suggests that flare stack reporting be included in the Onshore Petroleum and Natural Gas
Production source category. Many of BPs Alaska North Slope production operations have
stationary flares located at their onshore production facilities and to be complete, the flare source
type should be added. Therefore, BP recommends that flare stacks be added under 98.232 (c) (x).
BP suggests that EPA retain the option to use engineering calculations, company records or
similar estimates to compute volumetric flow. This same approach should be applied to the
inclusion of flare stacks in Onshore Petroleum and Natural Gas Production.
Alaska North Slope Operations Source Category Classification
BP believes that the proper classification of the North Slope Alaska oil production operations,
including the handling and use of associated natural gas, is Onshore Petroleum and Natural Gas
Production. Although the Technical Support Document seems to support this conclusion, the rule
is not completely clear and BP requests that EPA specifically clarify this uncertainty. Despite the
large amount of associated gas which is handled and treated on a daily basis, the North Slope
operations do not produce gas or NGLs for sales. The sole purpose is production of crude oil
which supports the classification wholly within the Onshore Petroleum and Natural Gas
Production source category.
Due to the comprehensive coverage of source types and the basin construct in the Onshore
Petroleum and Natural Gas Production source category, classifying all of the North Slope crude
oil production operations within this source category will not exempt significant emissions from
coverage. BP requests that EPA clarify that all Alaska North Slope operations properly classify
within the Onshore Petroleum and Natural Gas Production Category.
Response: EPA agrees with the commenter regarding flare stacks and todays final rule states
that flare stacks are one of the emissions sources to be monitored and reported in the onshore
petroleum and natural gas production segment. Todays final rule does allow use of engineering
calculations based on process knowledge, company records or similar estimates to compute
volumetric flow and composition of gas to flares.
EPA does not agree on grouping the natural gas processing facility on the North Slope of Alaska
with onshore petroleum and natural gas production. In developing the petroleum and natural gas
industry segments, EPA considered the major sources and how to gather emissions data in the
most cost-effective manner. Natural gas processing facilities, such as that on the North Slope of
Alaska, have very large, high pressure compressors, which are a significant source of emissions.
Todays final rule requires measurement methods to assess those emissions in processing
facilities. EPA determined that the size and widely dispersed nature of compressors in onshore
production did not warrant this same degree of emissions determination, and in todays final rule,
is allowing the use of equipment count and emissions factors for onshore production
compressors. To designate the North Slope of Alaska processing facility with onshore production
would not achieve the emissions data that EPA needs to inform future policy.
1647

EPA has proposed defining a new term, "onshore petroleum and natural gas production facility ."
See 75 Fed. Reg. 18,608 at 18,647 (Apr. 12, 2010). This definition is not consistent with
industry's and regulators' existing interpretation of the term "facility" in the oil and gas ("O&G")
upstream sector under the Clean Air Act ("CAA"). Therefore, Chesapeake recommends that if
EPA plans to aggregate exploration and production sites, EPA should use a different term, such
as a "threshold reporting area" or "roll-up area threshold" to define the geographical area that
upstream oil and gas facilities will use to assess the Mandatory Reporting Rule ("MRR")
applicability threshold. Utilizing a different term for this geographic area would help alleviate
any confusion that might arise from the use of the word "facility" as used in other sections of the
CAA.
If EPA adopts Chesapeake's suggested terminology, the definition of "onshore petroleum and
natural gas production facility" would need to be modified to read as follows: "Onshore
petroleum and natural gas production reporting area means all petroleum or natural gas facilities
and equipment associated with all petroleum or natural gas production under common ownership
or common control by an onshore petroleum and natural gas production owner or operator
located in a single hydrocarbon basin as defined. . . . .. 40 C.F.R. 98.238.
If EPA adopts this terminology, the term "reporting area" would also need to be incorporated
into the definition of the source category of onshore petroleum and natural gas production as
follows: "Onshore petroleum and natural gas production . Onshore petroleum and natural gas
production reporting area means all facilities associated with the production of petroleum or
natural gas including but not limited to. . . ." 40 C.F.R. 98 .230(a)(2) .
Response to Comments and response to EPA-HQ-OAR-2009-0923-1044-1. EPA does not agree
with the use of the term reporting area instead of facility. For further details, please see the
response to EMAIL-0001-1 (comment also located in rulemaking memo Early Comment
Submissions in docket EPA-HQ-OAR-2009-0923).

Commenter:Karen Ritter
Second, API concurs with EPAs statement that, quote, the facility definitions proposed in this
rule do not impact requirements under other EPA regulations, for example, new source review,
end of quote. While here in Subpart W, there may be a need to cluster equipment in activities for
1648
reporting, the terms used in conjunction with the facility definitions under Subpart W should not
impact any current or future aggregation determinations
Response: As regards impact of this rule on other EPA regulations, please see response to EPA
14.1.1
BASIN
Commenter:

Upstream Production Sites Should Not Be Aggregated
EPA should not require aggregation of upstream production sites. This imposes a
disproportionate burden on our industry by requiring data collection, recordkeeping, and
reporting for hundreds of thousands of operated sites. As the rule is written, even very small
operators will have to calculate and aggregate GHG emissions to determine if they must report or
not report.
EPA states in the preamble that the information gathered by this rule would inform and are
relevant to EPAs carrying out a wide variety of Clean Air Act (CAA) provisions. It is
understood that the GHG emission inventory data will be used to define future permitting and
compliance programs. For this reason, we do not agree with redefining the term facility in this
part. If the information in this program will be used for other CAA programs, then it must be
collected in a manner that is consistent with how it will be applied in those programs.
Future GHG permitting and compliance programs under the CAA will still rely on the definition
of facility as found in 40 C.F.R. 52.21(b)(6):
Building, structure, facility, or installation means all of the pollutant-emitting activities which
same industrial grouping if they belong to the same Major Group (i.e., which have the same
first two digit code) as described in the Standard Industrial Classification Manual, 1972, as
amended by the 1977 Supplement (U. S. Government Printing Office stock numbers 41010066
and 003005001760, respectively).
In accordance with this definition and multiple case-by-case aggregation determinations by EPA
and the states, most production operations across a single hydrocarbon basin cannot be
aggregated into one facility. If these production operations are not aggregated by basin, then each
facility or well site must individually be assessed for applicability to the future permitting and
1649
compliance programs. At this time, an applicability threshold of 25,000 metric tons CO2e or
more per year has been proposed for permitting, and that threshold will apply at the facility level.
If the purpose of this rule is to inform future rule-making, there is no benefit to collecting
information for an aggregation of many small facilities that will not be subject to the future rules.
We request that this definition be modified to be consistent with the definition employed in other
CAA programs.
Response: To reduce burden on the industry, EPA plans to develop screening tools to assist in
the determination of which facilities are required to report under todays final rule; please see
Section II.F of the preamble to todays final rule for further details. This will avoid the burden to
non-reporters in determining whether or not to report.
For further details, please see Topic 1: GHG Reporting under Subpart W and the Consolidated
2009-0923-1044-1, EPA-HQ-OAR-2009-0923-1044-6, and EPA-HQ-OAR-2009-0923-1005-2.
Commenter:

Basin Level Aggregation
EPA is seeking comment on whether to use basin level for aggregation as proposed, or field
level. If EPA goes forward with aggregation of upstream production sites, a field level
approach would be much more practical for the following reasons:
Some basins are very large (the Permian Basin stretches from southeast New Mexico and
southwest Texas nearly to Louisiana. Basins can typically cross state lines such that GHG
reporting by basin would not allow states or the EPA to determine which state the emissions are
from. In addition, basins can cross multiple regional and district operating boundaries for
companies making gathering of emissions information data more difficult.
EPA has overestimated the reduction in emissions coverage for field level aggregation versus
basin level aggregation. Most large and small operators concentrate in producing from certain
fields within a basin. The Barnett Shale in Texas covers 24 counties. However, one large
independent operates approximately 2,000 wells, but with 95% of those in just three or four
counties.
The quality of emission estimates would be greater for emissions estimates made on a field
level rather than a basin level. Oil and gas characteristics and the method of separation and
processing are directly related to the field from which the oil and gas is produced. Fields are well
defined by state oil & gas commission rules. When wells are comingled, the percent of
production from a particular field (sand or formation) can be assessed using royalty calculations
to provide the percentage production from a particular field. The field which produces the
majority of the oil and gas for a comingled well can be used as the field that a well reports under.
1650
Aggregation by basin level would complicate efforts for simple auditing by EPA as it would be
difficult to determine whether emissions reported are reasonable compared to emissions reported
from other companies. EPA would not know how much production came from various fields
within that basin.
For all the above reasons, we support the APIs proposed approach that would allow operators
the flexibility to report at the basin level but allow them to identify rational reporting units
within a basin that have similar operations, equipment, fluid properties, and physical properties,
and enable the use of population average emission factors and approaches for emission
quantification. This would allow calculation and reporting of emissions from reasonable well
groupings such as known fields within a basin or even by county within a basin.
Response: EPA does not agree with the commenter on the reasons for using a field level facility
definition. With regards to the size of the basin, please see response to EPA-HQ-OAR-2009
0923-1014-4 for further details. Furthermore, basins are more representative of characteristics of
emissions and not State boundaries.
EPA does not agree with the commenter on the differences between field and basin level
coverage. The commenter has provided an example that cannot be applied across the country. Oil
and gas operations vastly vary in characteristics across the country and cannot be analyzed using
specific case scenarios like the one provided by the commenter. Therefore, EPA has used actual
data across the country to estimate emissions coverage, which is the mostly reasonable way of
conducting the analysis. Please see Section 5 of the TSD for further details.
EPA requires the monitoring of certain emissions sources, e.g. well venting sources, by field so
as to capture the variation in emissions patterns based on the characteristics of the field. The
magnitude of emissions from other sources such as pneumatic devices do not depend on the type
of formation, hence a field level reporting is irrelevant. Therefore, EPA does not find an issue
with basin level reporting. As regards auditing, a basin level approach is in fact helpful to EPA
since the delineation is based on county boundaries, which are easy to follow.
Finally, EPA does not agree with the use of the term reporting units instead of facility. For

Organization: EC/R Incorporated
Commenter: Robert Vogt
I was wondering if you could please tell me where to find a map of the basins specified in the
GHG reporting rule for Subpart W.
1651
In particular, the rule documents the use of the American Association of Petroleum Geologists
(AAPG) classifications of a basin boundaries. Is there a map available that shows these basin
boundaries? If so, could you please let me know where it can be found?
Response: The AAPG classification of basins is available as a list of codes and map from AAPG
and the details on where they can be found are provided in Section 98.7 of todays final rule
covering standardized methods incorporated by reference.

Organization: Texas Commission on Environmental Quality
Commenter: Mark R. Vickery
The term "basin" is not defined in the rule language but is discussed in the preamble. The TCEQ
recommends that the definition of basin, because of its importance in the definition of a facility
for the onshore petroleum and natural gas production sector reporting, be added to the rule.
Response: EPA agrees with the commenter. In todays final rule, EPA has provided a separate
definition for basin. Please see Section 98.238 Definitions of todays final rule.

Furthermore, IOGA-WV notes that the size of the AAPG's designated hydrocarbon basins varies
dramatically across the country. Indeed, the entire state of West Virginia is divided into only two
AAPG hydrocarbon basins-the "Appalachian basin" (160) and the "Appalachian basin (Eastern
Overthrust area)" (l60A)-both of which span hundreds of miles along the entire Appalachian
region, from New York down to Tennessee, Alabama and Georgia. In contrast to the vast
hydrocarbon basins that encompass the Appalachian region, many of the AAPG basins in the
central and western parts of the country are a mere fraction of this size. Relying on the AAPG's
delineations thus imposes a significantly disparate impact on individual oil and gas operators
depending on the geographic region in which their operations are located. The substantial
variation in the size of the AAPG basins effectively eliminates them as an appropriate basis for
determining the applicability of a nationwide emissions reporting scheme
Response: EPA does not agree with the commenter. In any facility definition, either defined
using physical or geographical boundaries, there are always facilities of various sizes. This is the
case with all industries, not just oil and gas. Hence, EPA does not consider the varying basin size
as defined by AAPG as an issue and retains this definition for the purpose of this rulemaking.
For further details on why EPA chose the AAPG basin definitions, see Section II. D of the
1652

Commenter: Kim Dang
Reporting of Onshore Production Emissions at Basin Level. All wells within a producing basin
should not be classified as a single onshore production facility. 247 National Emissions
Standards for Hazardous Air Pollutants (NESHAPS), 248 and New Source Review (NSR) have
always defined a stationary source or facility as a discrete equipment site, such as an individual
wellhead. 249 Indeed, Section 112 of the CAA explicitly prohibits the aggregation of multiple oil
and gas wells, including those under common ownership and control, for purposes of
determining applicability of hazardous air pollutant standards. 250 Introducing a different, and
counterintuitive, definition of facility for the sole purpose of GHG reporting can only lead to
confusion and inconsistency, given industry structure and long-established practice in other CAA
programs.
In addition, grouping all wells within a basin together for purposes of reporting makes little
conceptual sense. Basins tend to be extremely heterogeneous with respect to geology, fluid
composition, and equipment and operational practices; hence, they are typically subdivided by
operators into multiple fields or units. For example, all of Kinder Morgans production is within
the Permian Basin in West Texas, but Kinder Morgan maintains separate operations for four
different fields within the Permian Basin, including the SACROC, Yates, Claytonville, and Katz
Strawn fields. Given the heterogeneity of basins, it will be difficult to draw any meaningful
policy conclusions from data collected at the basin level. Also, basins can typically cross state
lines such that GHG reporting by basin would complicate the attribution of emissions to
particular states.
Nevertheless, Kinder Morgan acknowledges EPAs concern that well-based reporting could
yield a large number of emissions reports, 251 depending on the reporting threshold that is
ultimately selected. To mitigate this potential problem, Kinder Morgan suggests that the basic
unit of reporting should be a wellhead, not a production basin; however, the number of reports
would be reduced by allowing reporting companies to group wellheads into coherent production
fields based on geographic features, operational characteristics, and other appropriate factors.
This approach would yield useful, policy-relevant data by grouping similar facilities together for
247
Proposed 40 C.F.R. SECTION 98.238.] Clean Air Act (CAA) regulations including NSPS, [Footnote 19: See
40 C.F.R. SECTION 60.3 (defining stationary source as any building, structure, facility, or installation which
emits or may emit any air pollutant). At 40 C.F.R. SECTION 52.21(b)(6), a building, structure, facility, or
installation is defined as all of the pollutant emitting activities which belong to the same industrial grouping, are
located on one or more contiguous or adjacent properties, and are under the control of the same person (or persons
under common control) . . . .
248
See 40 C.F.R. SECTION 63.760-777.
249
See 40 C.F.R. SECTION 52.01(a) (defining stationary source).
250
42 U.S.C. SECTION 7412(n)(4)(A).
251

1653
reporting purposes. At the same time, this approach would provide that the Mandatory Reporting
Rule does not disturb the conventional definition of stationary source that has been
consistently used in CAA programs, 252 and would allow for a manageable number of emission
report submittals for both industry members and EPA.
If EPA elects not to adopt this approach, Kinder Morgan also supports the concept of field-based
reporting, an approach EPA considers in the preamble to the proposed Subpart W. 253 Contrary to
EPAs observation that both [field and basin-level reporting] rely on geographical boundaries,
production fields under common operatorship provide far more coherent and manageable
reporting units than basins. As noted above, production basins can span multiple states and
thousands of wells with completely different characteristics. Production fields, by contrast, tend
to be geologically similar and contain wellheads that are comparable to each other and can be
sensibly aggregated. A 25,000 metric ton CO2-e threshold, if applied to production fields
designated by the Energy Information Administration, as EPA suggests in the preamble, would
yield more than adequate coverage of onshore petroleum and natural gas production emissions
while providing more useful data than basin-level reporting.
2009-0923-1005-2. EPA does not agree that basin level reporting will not provide meaningful
information. Please see response to EMAIL-0001-4 (comment also located in rulemaking memo
Early Comment Submissions in docket EPA-HQ-OAR-2009-0923) for further details.
EPA does not agree with the commenter on use of wellhead as a facility. Please see response to
EPA-HQ-OAR-2009-0923-1005-6, EPA-HQ-OAR-2009-0923-1005-2, and , and EPA-HQ
OAR-2009-0923-1305-46 for further details. As regards characteristics of different fields and
how they impact emissions, please see response to EMAIL-0001-4 (comment also located in
EPA does not agree with the commenter that the field level reporting would provide adequate
coverage of emissions. A field level reporting would result in about 28 percent less coverage
than basin level reporting. Please see Section 5 of the TSD for further details.

Organization : Independent Petroleum Association of Mountain States
Basin level reporting is not appropriate for natural gas and oil production sites. Defining a
facility to include many wells and associated equipment over a large area is contrary to the
252
In Section III.C of these comments, Kinder Morgan raises a similar concern with respect to the use of the term
facility in the proposed Subpart RR.
253

1654
Clean Air Act (CAA) definition and will subject many small companies to the reporting
requirements, contrary to EPAs stated goal of limiting impacts to small businesses.
Response to Comments. For EPA analysis on the impact on small businesses, please see Section
5.2 of the EIA to todays final rule.

EPA should define a simple and innovative system that balances the data collection and reporting
burden against the amount of emissions quantified. To achieve this simplification and balance,
API suggests that EPA allow operators to identify rational reporting units within a basin that
have similar operations, equipment, fluid properties and physical parameters and enable a
broader use of population average emission factors and approaches for emissions quantification.
API recognizes that the reporting threshold determination may still rest with the basin level
facility and the construct of the reporting units below the basin level will be implemented in such
a manner that they will cover all relevant operations of a given owner or operator within the
defined basin
Response: EPA does not agree with the commenter on the use of reporting units within a basin.
For further details, please see the response to EPA-HQ-OAR-2009-0923-1305-46.
14.1.2
FIELD

Commenter:
75 FR 18615: C. Definition of the Source Category, Onshore Natural Gas
Processing
Comment: WBIH recommends Field Level reporting for onshore petroleum and natural gas
production with a reporting threshold of 25,000 mtCO2e.
EPA is seeking comments on advantages to require reporting at the field level instead of the
basin level. The purpose of the Energy Information Administration Oil and Gas Field Code
Master publication is to provide standardized names and codes for identifying domestic fields.
"Use of these field names and codes fosters consistency of field identification by government
1655
and industry. As a result of their widespread adoption they have in effect become a national
standard."
Referencing the discussion regarding the combination of direct measurement and engineering
estimation, specifically related to flaring and unconventional well drilling, EPA stated "As a
result, EPA proposes the development of a field-specific emission factor either by direct
measurement of flow rate of hydrocarbons using a meter or by and engineering estimation based
on choke pressure drop." Field level reporting provides the opportunity to use similar gas and oil
compositions for emissions quantifications and supports EPA desire to develop field-specific
emissions factors, which is more credible data for development of GHG regulatory programs.
One size does not fit all
Response: EPA does not agree with the commenter and is retaining the basin level facility
definition in todays final rule. . The AAPG basin definition is as standardized as the EIA Field
Code Master, so there is no distinct advantage in terms of standardization in using the field over
the basin as a facility definition. In addition, section of the preamble to the April 2010 proposed
rule referenced by the commenter refers specifically to well venting sources where sampling
within a field has been allowed as an option. EPA notes that this is in no way related to other
emissions source types where such a sampling option is not provided in the rule. Also, the well
venting sampling option is provided for fields within a reporting basin.

First, we believe you cast your net too broadly when defining the concept of facility for natural
gas distribution. As a result, the proposed rule would likely apply to many more gas utilities and
require a much greater burden than you estimate. It is not clear to us whether you intend to
define the facility as each distribution system or as all of the distribution systems owned by a
single Distribution Company. The two are not aligned all the time. If you plan to combine all
of the separate distribution systems of a single company in one facility that spans hundreds of
communities if not thousands across a state some of our companies span several states as well
you would have many more utility companies that would be likely to exceed your 25,000 ton
per year threshold for reporting.
Response: EPA disagrees that it cast the net too broadly with the definition of facility for natural
gas distribution. Please see the response to comment EPA-HQ-OAR-2009-0923-1016-26 for
further details.
14.2
DEFINITION OF NATURAL GAS
1656

Aka strongly recommends that the definition in Section 98.230(a)(3) be revised to exclude field
gathering and boosting stations from the Onshore Natural Gas Processing source category
definition, and that a new source category definition be included for Onshore Natural Gas
Gathering Compression. Using the definition of Onshore Natural Gas Transmission as a
template, AKA respectfully requests that the source category definition for Onshore Natural
Gas Gathering Compression be changed to the following:
Onshore natural gas gathering compression means any physically adjacent combination of
compressors that move natural gas from production fields or other compression facilities into
natural gas processing facilities, other gathering compression facilities, transmission pipelines,
storage facilities, or other end users. In addition, natural gas gathering compressor facilities may
include equipment for liquids separation, natural gas dehydration, and tanks for the storage of
water and hydrocarbon liquids.
Response: EPA has not included field gathering and boosting stations from the Onshore Natural
Gas Processing source category in todays final rule. For further information, please see Section
II.F of the preamble to todays final rule.

The Sour Natural Gas definition states that it is natural gas that contains significant
concentrations of hydrogen sulfide and/or carbon dioxide that exceed the concentrations
specified for commercially saleable natural gas. The concentration of carbon dioxide does not
determine whether natural gas is considered sour. The oil and gas industry considers natural
gas to be sour based only on the gas hydrogen sulfide concentration. Although carbon dioxide
content is indirectly controlled by heat content specifications, most natural gas contracts do not
contain a specification for carbon dioxide content unless the natural gas contains an excessive
concentration of carbon dioxide. In addition, the concentration of hydrogen sulfide at which a
natural gas stream is considered sour is not a standard value across the industry. This
concentration varies according to the production area and the sales contract. We request that the
sour natural gas definition not reference carbon dioxide content, and we suggest that EPA
define the concentration of hydrogen sulfide at which natural gas is consistently considered
sour to be at least 100 ppmv.
The Sweet Gas definition states that it is natural gas with low concentrations of hydrogen sulfide
and/or carbon dioxide. As discussed in the comment for sour natural gas, the concentration of
1657
carbon dioxide does not determine whether natural gas is considered sour or sweet. We
request that the sweet gas definition not reference CO2 content.
Response: EPA has reviewed these comments and disagrees that any composition conditions
should be placed on the definitions of sour or sweet natural gas. The reporting requirement is
related to gas feed to an acid gas removal unit and gas that is suitable for transmission,
distribution and sales. The terms as used in subpart W have no effect other than to distinguish
between gas suitable with regard to H2S and CO2 content for sales and gas requiring acid gas
removal.
14.3
ONSHORE GAS PROCESSING BOUNDARY (PRODUCTION VERSUS

PROCESSING)
Commenter:

Defining natural gas gathering system compression facilities as part of a gas processing plant is
inappropriate.
The definition of a gas processing plant is well established in the context of the Clean Air Act
and it does not incorporate stand alone gathering system compression facilities. Gas compression
facilities, whether in the gathering systems upstream of gas processing or in the natural gas
transmission sector, are discreet and easily identifiable facilities. These facilities have been
treated as stand alone sites in all previous applications of the Clean Air Act, including all
permitting programs and for GHG reporting under 40 CFR Part 98 Subpart C. Continuing this
historical treatment of these facilities is not only the most appropriate action, it will also
eliminate the certain confusion that will result from a new definition solely for use in Subpart W
of Part 98.
Even the EPA appropriately recognizes the treatment of gathering system compressor stations as
stand alone facilities. On page 26 of the Background Technical Support Document for the
Petroleum and Natural Gas Industry, the EPA states:
Both gathering/boosting stations and natural gas processing facilities have a well defined
boundary within which all processes take place.
Further, on page 18613 of the Subpart W preamble the EPA states that,
For some segments of the industry (e.g., onshore natural gas processing facilities, natural gas
transmission compression facilities, and offshore petroleum and natural gas facilities),
lend themselves to identifying scope of reporting and responsible reporting entities.
1658
In direct conflict to the above statements, the proposed Subpart W definition of Onshore Natural
Gas Processing in 98.230(3) appears to require GHG emissions from gathering compression
facilities and the gathering pipelines be rolled-up with gas processing plant emissions by stating
(emphasis added):
In addition, field gathering and/or boosting stations that gather and process natural gas from
multiple wellheads, and compress and transport natural gas (including but not limited to
flowlines or intra-facility gathering lines or compressors) as feed to the natural gas processing
plants are considered a part of the processing plant. Gathering and boosting stations that send the
natural gas to an onshore natural gas transmission compression facility, or natural gas
distribution facility, or to an end user are considered stand alone natural gas processing facilities.
All residue gas compression equipment operated by a processing plant, whether inside or outside
the processing plant fence, are considered part of natural gas processing plant.
GPA strongly recommends the above text in 98.230(3) be deleted from the Onshore Natural
Gas Processing source category definition and a new source category definition be included for
Onshore Natural Gas Gathering Compression. Using the definition of Onshore Natural Gas
Transmission as a template the source category definition for Onshore Natural Gas Gathering
Compression should be:
Onshore natural gas gathering compression means any fixed combination of compressors that
move natural gas from production fields or other compression facilities into natural gas
processing facilities, other gathering compression facilities, transmission pipelines, storage
facilities, or other end users. In addition, natural gas gathering compressor facilities may include
equipment for liquids separation, natural gas dehydration, and tanks for the storage of water and
hydrocarbon liquids.
GPA also suggests that the same GHGs and sources reported by processing plants under
98.232(d) be reported by gathering compression facilities.
In addition to the conflicts created by the proposed definition in Part 98, Subpart W regarding
applicability of gathering compression facilities, the following issues arise from the proposed
Subpart W definition of Onshore Natural Gas Processing in 98.230(3).
a) Increases the number of facilities subject to Subpart C for combustion GHG emissions by
approximately 15 times, with no consideration for the size of each individual facility. This results
from the fact that, numerically, there are typically 10-20 gathering compression facilities for
every gas processing plant.
b) Gathering compression facilities are often not uniquely associated with a single gas plant,
many gathering compression facilities have connections to multiple gas plants and gas is routed
to a plant based on business needs and conditions. Associating any one gathering compression
facility to an individual processing plant is not always possible.
c) Rolling-up gathering compression facility emissions with processing plant emissions is in
1659
conflict with EPAs assessment of the impact of the proposed Subpart W. EPA inaccurately
states in the proposed Subpart W preamble on page 18616,
there are a reasonable number of reporters. Most natural gas processing facilities proposed
for inclusion in this supplemental proposed rulemaking would already be required to report
under subpart C and/or subpart NN of the Final MRR.
In fact, this proposal will increase the number of facilities subject to Subpart C for compression
GHG emissions by approximately 15 times. One GPA member company that has about 45
facilities subject to Subpart C reporting for 2010 would have over 700 facilities subject to
Subpart C as a result of this proposal. This level of reporting with no consideration for individual
facility emission levels is neither reasonable nor appropriate.
d) EPA makes the following conclusion on page 18619 in the preamble to the proposed Subpart
W,
requiring only a small fraction of total facilities to report.
appropriate.
e) Gathering compression facilities are typically small facilities and widely dispersed, many of
these compression facilities have only 1 or 2 compressors. Significantly sized facilities are
already subject to Subpart C reporting if the combustion emissions exceed 25kTon/year and
individual compression facilities that exceed 25kTons/year including Subpart W emissions
would also begin reporting as required. The proposed Subpart W, however, treats these small and
widely dispersed facilities as if they are the same size and complexity of a gas processing plant.
Examples of proposed requirements that may be appropriate for processing plants but unduly
burdensome for small and widely dispersed gathering compression facilities include: collecting
extensive data and modeling tank emissions, compressor rod packing vents, leak detection using
optical imaging or population factors, and quarterly sampling of gas streams.
Response: EPA has decided not to include natural gas gathering and boosting in todays final
rule. For more information, please see Section II.F of the preamble to todays final rule. As a
result of excluding natural gas gathering and boosting, EPA has redefined onshore natural gas
processing and onshore petroleum and natural gas production to reflect this change. EPA has
also clarified the boundary between onshore processing and transmissions compressor stations.
For further details, please see response to EPA-HQ-OAR-2009-0923-1004-14. However EPA
disagrees, and continues to include residue gas compressors in the gas processing facility, that
are owned and operated by a gas processing facility, and dedicated to moving sales gas from that
gas processing facility to a transmission system, whether inside or outside the processing facility
fence line. Residue gas compressors are large sources of emissions and move a significant
output of the processing facility, and therefore are included in gas processing. Please see
1660

PAW believes that it is unclear when Production is required to include gathering, and when
boosting/processing includes gathering. PAW reiterates that it is unclear in the rule where
Production site responsibility ends and Gathering/Processing site responsibility begins.
todays final rule. Furthermore, EPA has clarified the boundaries of onshore production as it
relates to other segments. For further details, please see Section 98.2 of todays final rule.
98.230(a)(3): Onshore Natural Gas Processing Plant Source Definition. In NSPS KKK, a natural
gas processing plant is defined as, any processing site engaged in the extraction of natural gas
liquids from field gas, fractionation of mixed natural gas liquids to natural gas products, or both.
Nonfractionating plant means any gas plant that does not fractionate mixed natural gas liquids
into natural gas products. EPAs treatment of onshore natural gas processing plants does not
appear to align with the NSPS KKK definition of natural gas facility. For purposes of
applicability, a gathering/boosting station without gas processing is not subject to the
requirements of KKK. Therefore, a gathering/boosting station that is not owned by a processing
plant should not be considered part of that processing plant facility.
boosting segment of the industry. For further details, please see Section II.F. of the preamble to
todays final rule.
98.232(c)(9): Gathering pipeline fugitives. PAW believes that it is unclear when Production is
required to include gathering, and when boosting/processing includes gathering. PAW reiterates
that it is unclear in the rule where Production site responsibility ends and Gathering/Processing
site responsibility begins.
1661
relates other segments. For further details, please see Section 98.2 of todays final rule.
In Subpart W 98.230(a)(3) Onshore Natural Gas Processing Plant Source Definition, PAW
believes that this definition overlaps with processing plants, which also include gathering lines. It
is unclear in the rule where Production site responsibility ends and Gathering/ Processing site
responsibility begins. Furthermore, the rule does not specifically imply that gathering and
boosting leading to a gas plant must be under common ownership/control of the gas plant in
order to be included with the gas plants GHG emissions. Furthermore, it is not explicitly clear
that the associated equipment (e.g., upstream compression and gathering) of gas processing
facilities must be under common ownership and control. EPA should clarify this point.
1004-14.
Commenter: Kim Dang
EPA should create an onshore natural gas gathering source category that addresses natural gas
gathering facilities in the same manner as natural gas transmission facilities. Neither the industry
nor EPA in other Clean Air Act programs has treated gathering facilities as part of onshore
production facilities or processing plants.
Response: Todays final rule does not include gathering and boosting sector. For further details,
please see Section II.F of the preamble to todays final rule.
Natural Gas Processing Facility Definition
The proposed definition of natural gas processing facilities includes "gathering and boosting
stations that dehydrate and compress natural gas to be sent to natural gas processing facilities or
directly to natural gas transmission or distribution systems." The definition of a gas processing
1662
plant is well established in the context of the Clean Air Act and it does not incorporate stand
alone gathering system compression facilities. Gas compression facilities, whether in the
gathering systems upstream of gas processing or in the natural gas transmission sector, are
discreet and easily identifiable facilities. These facilities have been treated as stand-alone sites in
all previous applications of the Clean Air Act, including all permitting programs and for GHG
reporting under 40 CFR Part 98 Subpart C. Continuing this historical treatment of these facilities
is not only the most appropriate action, it will also eliminate the certain confusion that will result
from a new definition solely for use in Subpart W of Part 98.
Even the EPA appropriately recognizes the treatment of gathering system compressor stations as
stand- alone facilities. On page 26 of the Background Technical Support Document for the
Petroleum and Natural Gas Industry, the EPA states:
"Both gathering/boosting stations and natural gas processing facilities have a well defined
boundary within which all processes take place".
Further, on page 18613 of the Subpart W preamble the EPA states that,
"For some segments of the industry (e.g., onshore natural gas processing facilities, natural gas
lend themselves to identifying scope of reporting and responsible reporting entities. "
In direct conflict to the above statements, the proposed Subpart W definition of Onshore Natural
Gas Processing in 98.230(3) appears to require GHG emissions from gathering compression
facilities and the gathering pipelines be rolled-up with gas processing plant emissions by stating
(emphasis added):
"In addition, field gathering and/or boosting stations that gather and process natural gas from
the processing plant fence, are
considered part of natural gas processing plant. "
Dominion strongly urges that the above text in 98.230(3) be deleted from the Onshore Natural
1663
Gas Processing source category definition and a new source category definition be included for
"Onshore Natural Gas Gathering Compression" because gathering facilities are stand alone
facilities and should be treated as separate facilities for applicability purposes. Using the
definition of Onshore Natural Gas Transmission as a template the source category definition for
"Onshore Natural Gas Gathering Compression" should read:
"Onshore natural gas gathering compression means any fixed combination of compressors that
move natural gas from production fields or other compression facilities into natural gas
processing facilities, other gathering compression facilities, transmission pipelines, storage
facilities, or other end users. In addition, natural gas gathering compressor facilities may include
equipment for liquids separation, natural gas dehydration, and tanks for the storage of water and
hydrocarbon liquids. "
Subpart W definition of Onshore Natural Gas Processing in 98.230(3). These issues would be
resolved by creation of a specific source category for onshore natural gas gathering compression.
a) Increases the number of facilities subject to Subpart C for combustion GHG emissions by
approximately 15 times, with no consideration for the size of each individual facility.
This results from the fact that, numerically, there are typically 10-20 gathering compression
facilities for every gas processing plant.
b) Gathering compression facilities are often not uniquely associated with a single gas plant,
many gathering compression facilities have connections to multiple gas plants and gas is routed
to a plant based on business needs and conditions_ Associating anyone gathering compression
facility to an individual processing plant is not always possible.
c) Rolling-up gathering compression facility emissions with processing plant emissions is in

conflict with EPA's assessment of the impact of the proposed Subpart W. EPA inaccurately
states in the proposed Subpart W preamble on page 18616,
" ... there are a reasonable number of reporters. Most natural gas Processing facilities proposed
under subpart C and/or subpart NN of the Final MRR."
In fact, this proposal will dramatically increase the number of facilities subject to Subpart
C for compression GHG emissions by approximately 15 times, essentially increasing coverage of
gas gathering and processing facilities to nearly 100% regardless of facility size, which is unduly
1664
burdensome and neither reasonable nor appropriate.
d) Gathering compression facilities are typically small facilities and widely dispersed. Many of
these compression facilities have only I or 2 compressors. Significantly sized facilities are
already subject to Subpart C reporting if the combustion emissions exceed
25k Ton/year and individual compression facilities that exceed 25kTons/year including
Subpart W emissions would also begin reporting as required. The proposed Subpart W, however,
treats these small and widely dispersed facilities as if they are the same size and
complexity of a gas processing plant. Examples of proposed requirements that may be
appropriate for processing plants but unduly burdensome for small and widely dispersed
gathering compression facilities include: collecting extensive data and modeling tank emissions,
measuring leaks from compressor rod packing vents, leak detection using optical imaging or
population factors, and quarterly sampling of gas streams.
Response: Todays final rule does not include the gathering and boosting segment of the
industry. For further details, please see Section II.F of the preamble to todays final rule.
Including field gathering and/or boosting stations that gather and process natural gas from
multiple wellheads, and compress and transport natural gas as feed to processing plants is
inappropriate. Definition of this activity is well established in the context of the Clean Air Act,
which does not incorporate processing and compression into one operation. Incorporating these
operations into one reporting unit conflicts reporting requirements under other portions of this
Part, specifically 40 CFR Part 98, Subpart C. We recommend EPA separate these activities into
separate reporting units.
Commenter:
Source Type:
Gathering Pipeline Fugitives
40 CFR 98.233(r)
1665
Use Eq. W-19 to calculate emissions from all sources listed in 98.233(r) - see comment for listed
sources.
1. Total number of each type of emission source listed in 98.233(r)
2. For onshore petroleum and natural gas production facilities and onshore natural gas processing
facilities, concentration of GHGi (CH4 or CO2) in produced natural gas or feed natural gas; for
other facilities listed in 98.230 (b)(3) through (b)(8), GHGi equals one.
3. Total time the specific source associated with the fugitive emission was operational in the
2. If the tank vapors are continuous then use a meter (such as a turbine meter) to measure tank
vapors.
3. Use the appropriate gas composition from 98.233(u)(2)(iii) (i.e., the GHG mole percent in
transmission pipeline natural gas that passes through the facility for onshore natural gas
transmission compression facilities).
2. Use the calculation methodology of flare stacks in paragraph (n) of this section to determine
storage tank emissions from the flare.
Comments:
Yates Petroleum Corporation believes that it is unclear when Production is required to include
gathering, and when boosting/processing includes gathering. Yates Petroleum Corporation
reiterates that it is unclear in the rule where Production site responsibility ends and
Gathering/Processing site responsibility begins.
Commenter:
In Subpart W 98.230(a)(3) Onshore Natural Gas Processing Plant Source Definition, Yates
Petroleum Corporation believes that this definition overlaps with processing plants, which also
include gathering lines. It is unclear in the rule where Production site responsibility ends and
Gathering/ Processing site responsibility begins.
1004-14.
1666

The current definition of the "onshore natural gas processing plants" source
category is too broad. Proposed 98.230(a)(3) would define ''natural gas processing plant" to
include upstream components that should be considered separate from the plant itself, including
gathering lines and/or booster stations or compressor stations. Most importantly, upstream
compressor stations should not be considered as part of the onshore natural gas processing plant.
Emissions from upstream compressor stations should not be aggregated with emissions from
components that are properly considered as part of the plant. In practice, compressor stations are
considered to be separate from processing plants. Disaggregating such components from the
plant would be consistent with industry practice and would eliminate a significant problem with
the proposed language: the fact that upstream compressor stations are not always tied to or
associated with a single processing plant. There is a significant amount of interconnection
between various gas pipelines, compressor stations, and processing plants, such that a given
compressor station may be connected to and deliver gas to any number of different processing
plants. Accurately aggregating plant + compressor station emissions in this context would be
difficult or impossible.
Accordingly, the following sentence from 98.230(a)(3) should be revised as follows:
"[strike though: In addition, f] Field gathering and/or boosting stations that gather and process
natural gas from multiple wellheads, and compress and transport natural gas
[strike through: (including but not limited to flow lines or intra facility gathering lines or
compressors)] as feed to the natural gas processing plants should be considered
stand alone components for applicability under this source category, consistent
with applicable provisions in Subpart C [strike through: are considered a part of the processing
plant.]"
Another problem with 98.230(a)(3) is its provision that "[a]l1 residue gas compression
equipment operated by a processing plant, whether inside or outside the processing plant fence,
are considered part of natural gas processing plant." This provision should not be a blanket rule.
Equipment that is outside the plant's fence should be aggregated with the plant only if those
facilities are operating under the same Title V permit
1004-14 As regards residue compressors outside the fence line of the processing facility, EPA
does not find any issue with reporting emissions from them since the reporting requirement states
that the processing facility will only report residue compressors that they own or operate. This
data collection rule is not bound by Title V permitting requirements; please see response to EPA
1667
Commenter:
Comment on 98.230(a)(3): Onshore natural gas processing plants.
WBIH strongly recommends natural gas gathering and boosting stations be removed from the
onshore natural gas processing industry segment and included with onshore petroleum and
natural gas production industry segment. Natural gas gathering compressor stations and boosting
stations are considered "production field facilities" for current air quality permitting programs
and emission inventories
Commenter:
Comment: WBIH strongly recommends natural gas gathering and boosting stations be removed
from the onshore natural gas processing industry segment and included with onshore petroleum
and natural gas production industry segment
Natural gas gathering and boosting stations are considered "field production facilities" for
current air quality permitting programs and emissions inventories. The North American Industry
Classification System (NAICS) code for field production facilities is 211111 and the SIC code is
1311
The rule requires compressor stations to be included with processing plants. However,
compressor stations are frequently connected to more than one processing plant so that multiple
flow paths are possible. The proposed rule provides no guidance for how to address this
1668
situation. Complicating the matter further, multiple companies may own the compressor stations
connected to a plant. Again, the rule is not clear on how to handle reporting in such situations.
todays final rule.
Aka recommends that the source categories and definitions (98.236(a)) for onshore natural gas
production facility and onshore natural gas processing plant clearly define the difference
between the operations to extirpate overlap between the two.
Response: In todays final rule, EPA has clarified the boundaries of onshore production as it
1004-14 EPA has also clarified the definition of onshore natural gas processing facilities; please
see section II.F of the preamble to todays final rule for further details.
An onshore natural gas production facility may contain some of the processes that are currently
defined as part of an onshore natural gas processing plant (e.g., the removal of water, condensate,
and/or oil). Due to this confusion in the definitions of production and processing, many
production facilities could fall into both categories. This will result in different operators
defining their operations inconsistently, and will most likely result in facility emissions being
double counted in both source categories.
Response: In todays final rule, EPA does not require reporting of emissions from the gathering
and boosting segment of the industry. In addition, EPA has provided a facility throughput for
non-fractionation processing facilities below which they are not considered processing facilities
that will avoid any overlaps between onshore production and onshore processing facility
definitions; please see Sections II.D and II.E of the preamble to todays final rule. Finally, EPA
has clarified the boundaries of onshore production as it relates to other segments. For further
details, please see response to EPA-HQ-OAR-2009-0923-1004-14.
1669
Gathering compression facilities are often not uniquely associated with a single gas plant, and
many gathering compression facilities have connections to multiple gas plants, where gas is
routed to a plant based on business needs and conditions. Associating any one gathering
compression facility to an individual processing plant is not always possible.
todays final rule.
Section 98.233(r) Gathering pipeline fugitives
Onshore Production Source Category: Gathering pipeline fugitives
Section 98.232 (c) (9) lists gathering gas pipeline fugitives as a reportable source type for
onshore petroleum and natural gas production, however, Section 98.233 (r) includes gathering
gas pipeline fugitives only for the natural gas processing sector.
Section 98.233 (r) (3) Onshore natural gas processing facilities shall use the appropriate default
population emission factor listed in Table W-2 of this subpart for fugitive emissions from
gathering pipelines.
As discussed previously in Section I.C of this document, API recommends designating Onshore
Natural Gas Gathering and Collection Systems as a stand-alone sub-sector of the Onshore
Petroleum and Gas Production (OPGP) category, separate from Onshore Petroleum and Natural
Gas Production and Onshore Natural Gas Processing Plants. This provides a clear distinction
between production and processing operations, and eliminates duplicate reporting due to the
overlap of some operations and equipment.
For gathering compressor stations, API recommends that these not be grouped for the purpose of
threshold determination or reporting under Subpart W. Subpart C has already set the precedence
for the definition of a facility and a compressor station, regardless of it being for transmission
or gathering, the compressor station has been treated as a stand-alone facility under Subpart C.
API requests that EPA remain consistent with its Subpart C rule and regard each
gathering/boosting compressor station as an individual facility.
API also recommends that gathering pipelines be considered another separate reporting segment,
consistent with EPAs statements in its Background Technical Support Document (TSD). EPA
has acknowledged in its TSD that a separate reporting segment for gathering pipelines would be
an appropriate option: Unlike other segments of the petroleum and natural gas industry,
gathering systems may be owned by producers, processing plants, transmission companies, local
distribution companies, or independent gathering companies. Therefore, it is difficult to assign
1670
this portion of onshore production to one particular segment. One option is to require gathering
pipelines to be reported as an emission source. The other option is to have a separate segment
assigned to gathering pipelines. See Appendix F for further discussion on the options.
(reference TSD p.18).
Under Appendix F of the TSD, EPA further recognizes that gathering pipelines should be
reported as a separate segment at the company level. If GHG emissions from the natural gas
gathering pipeline segment were included, it would be most straightforward to have emissions
reported at the pipeline company level as this is consistent with the PHMSA reporting. TSD
p.108
industry. For further details, please see Section II.F of the preamble to todays final rule. EPA
has made appropriate adjustments to the monitoring methods and data reporting requirements per
this change in todays final rule.
IPANM contends that the 98.230(a)(3) Onshore Natural Gas Processing Plant Source definition
overlaps with processing plants, which also include gathering lines. It is unclear in the rule where
Production site responsibility ends and Gathering/ Processing site responsibility begins. The rule
does not clarify whether gathering and boosting lines leading to a gas plant must be under
common ownership/control of the gas plant in order to be included with the gas plants GHG
emissions. Furthermore, EPA must clarify what associated equipment such as upstream
compression and gathering of gas processing facilities must be under common ownership and
control in order to trigger the reporting requirements.
relates to other segments. For further details, please see Section 98.2 of todays final rule
The proposed rule includes a definition of "natural gas processing facility" that includes
"gathering and booster stations that dehydrate and compress natural gas to be sent to natural gas
processing facilities or directly to natural gas transmission or distribution systems." This
proposed definition is inconsistent with the existing definition of a gas processing facility
established in the context of the Clean Air Act which does not incorporate stand alone gather
1671
system compression facilities. These types of facilities have been treated as stand alone sources
in all previous applications of the Clean Air Act, including all permitting programs and for GHG
reporting under 40 CFR Part 98 Subpart, and when treated alone, in most cases such facilities are
so small that their. emissions never trigger permitting or reporting requirements. This definition
is of particular concern in West Virginia, where more than one company may be involved
between the production, processing, transmission and distribution of natural gas. In many cases,
where a gathering compression facility is not uniquely associated with a single gas plant,
associating anyone gathering compression facility to a individual processing plant is not always
possible. In these cases it would be very difficult to even locate such equipment in order to
inventory emissions. '
todays final rule and Topic 2: Aggregation of Gathering and Boosting Systems with Processing
Facilities in Volume 9 of the Response to Comments.
Alternatively, if EPA determines that GHG emissions from gathering lines must be reported, we
recommend that a new source category definition for gathering lines be included in the rule.
Such a definition should specify that gathering lines: (1) are located upstream of natural gas
processing plants; (2) carry produced gas; (3) do not include piping within gas plants,
compression facilities, and treatment facilities; and (4) do not include transmission lines, even if
located upstream of a gas plant.
Commenter: Jeff Applekamp:
inappropriate.
The definition of a gas processing plant is well established in the context of the CAA, and it does
not incorporate stand-alone gathering system compression facilities. Gas compression facilities,
whether in the gathering systems upstream of gas processing or in the natural gas transmission
sector, are discreet and easily identifiable facilities. These facilities have been treated as stand
alone sites in all previous applications of the CAA, including all permitting programs and for
greenhouse gas (GHG) reporting under 40 CFR Part 98 Subpart C. Continuing this historical
treatment of these facilities is not only the most appropriate action, it also will eliminate the
1672
certain confusion that will result from a new definition solely for use in Subpart W of Part 98.
EPA has repeatedly recognized the treatment of gathering system compressor stations as stand
alone facilities in past regulations and other documents, including in its Background Technical
Support Document for the Petroleum and Natural Gas Industry, which states:
U.S. Environmental Protection Agency, Fugitive Emissions Reporting from the Petroleum and
Natural Gas Industry: Background Technical Support Document (TSD) at 20
In the proposed Subpart W preamble, EPA states:
identifying the facility is clear since there are physical Air and Radiation Docket and Information
Center boundaries and ownership structures that lend themselves to identifying scope of
reporting and responsible reporting entities.
75 Fed. Reg. at 18613.
The term natural gas processing plant has been previously and consistently defined in other
EPA programs. A natural gas processing plant (gas plant) is defined under Standards of
Performance for Equipment Leaks of VOC from Onshore Natural Gas Processing Plants as any
processing site engaged in the extraction of natural gas liquids from field gas, fractionation of
mixed natural gas liquids to natural gas products, or both. See 40 CFR Part 60, Subpart KKK.
Subpart KKK also defines natural gas liquids as the hydrocarbons, such as ethane, propane,
butane, and pentane, that are extracted from field gas. Accordingly, natural gas liquids are
those light-end hydrocarbons that are extracted from a natural gas stream that must be stored in a
closed pressurized tank to be maintained as a liquid. Taken together, these definitions do not
incorporate the removal of water, condensate, oil, H2S, CO2, or any other component other than
natural gas liquids.
Additionally, natural gas processing plant (gas plant) is defined under the Chemical Accident
Prevention program as any processing site engaged in the extraction of natural gas liquids from
field gas, fractionation of mixed natural gas liquids to natural gas products, or both, classified as
North American Industrial Classification System (NAICS) code 211112 [previously Standard
Industrial Classification (SIC) code 1321]. See 40 CFR Part 68.
In direct conflict with the foregoing statements by EPA and existing regulatory definitions, the
proposed Subpart W definition of Onshore Natural Gas Processing in Section 98.230(a)(3) of the
proposed rule appears to combine GHG emissions from gathering compression facilities and the
gathering pipelines with gas processing plant emissions:
1673
75 Fed. Reg. at 18636 (emphasis added).
The oil and gas industry commonly understands the term natural gas processing plant to mean
the extraction of natural gas liquids from a natural gas stream to meet the heat content
specification of sales contracts. A Joule-Thompson unit or a refrigeration unit typically Air and
Radiation Docket and Information Center completes this natural gas liquids extraction. It is not
appropriate to introduce a different definition for natural gas processing plant in this program.
The contrast between industrys longstanding usage of the term and other EPA regulations with
the proposed definition in Subpart W will lead to confusion among owners and operators, and
will be inconsistent with other existing EPA regulations under the Clean Air Act.
Accordingly, GPA strongly recommends that the definition in Section 98.230(a)(3) be revised to
exclude field gathering and boosting stations, including any associated acid gas treating facilities,
from the Onshore Natural Gas Processing source category definition, and that a new source
category definition be included for Onshore Natural Gas Gathering Compression and Treating
Facilities. Using the definition of Onshore Natural Gas Transmission as a template, GPA
respectfully requests that the source category definition for Onshore Natural Gas Gathering
Compression and Treating Facilities be changed to the following:
Onshore natural gas gathering compression and treating facilities means any physically adjacent
combination of compressors that move natural gas from production fields or other compression
facilities into natural gas processing facilities, other gathering compression facilities,
transmission pipelines, storage facilities, or other end users. In addition, natural gas gathering
compression and treating facilities may include equipment for liquids separation, natural gas
dehydration, acid gas removal, and tanks for the storage of water and hydrocarbon liquids. These
facilities do not include equipment designed to extract natural gas liquids.
Subpart W definition of Onshore Natural Gas Processing in 98.230(a)(3).
1004-14.
1674

The proposed rule increases in the number of facilities subject to Subpart C for combustion GHG
emissions by approximately 15 times, with no consideration for the size of each individual
facility. This results from the fact that, numerically, there are typically 10-20 gathering
compression facilities for every gas processing plant.
b) Gathering compression facilities are often not uniquely associated with a single gas plant, and
many gathering compression facilities have connections to multiple gas plants, where gas is
routed to a plant based on business needs and conditions. Associating any one gathering
compression facility to an individual processing plant is not always possible.
Rolling-up gathering compression facility emissions with processing plant emissions is in
conflict with EPAs assessment of the impact of the proposed Subpart W. EPA inaccurately
states in the proposed Subpart W preamble:
[T]here are a reasonable number of reporters. Most natural gas processing facilities proposed
under subpart C and/or subpart NN of the Final MRR.
75 Fed. Reg. at 18616.
In fact, this proposal will increase the number of facilities subject to Subpart C for compression
GHG emissions by approximately 15 times. One GPA member company that has about 45
facilities subject to Subpart C reporting for 2010 would have over 700 facilities subject to
Subpart C as a result of this proposal. This level of reporting with no consideration for individual
facility emission levels is neither reasonable nor appropriate.
d) EPA makes the following conclusion in the preamble to the proposed Subpart W:
[The rule would] requir[e] only a small fraction of total facilities to report.
75 Fed. Reg. at 18619.
1675
appropriate.
Section 98.230(a)(3): The Onshore natural gas processing plants source category states that those
plants are designed to separate and recover natural gas liquids (NGLs) or other non-methane
gases and liquids from a stream of produced natural gas, including oil and condensate removal,
water removal, separation of natural gas liquids, sulfur and CO2 removal, fractionation of NGLs,
or other processes, and also the capture of CO2. This definition is too broad and captures
processes that occur throughout the entire natural gas production and processing system, not just
at a gas processing plant. In particular, the phrase oil and condensate removal includes all
production separators, and the phrase water removal includes dehydration units, free water
knockouts, separators, skim tanks, etc. These types of units can exist anywhere from the
production well site, through processing and transmission, to immediately before distribution.
Response: EPA does not agree with this interpretation of subpart W. All segments of the
Petroleum and Natural Gas Systems rule with exception of onshore production and natural gas
distribution generally conform to the definition of a facility in Section 98.6 of subpart A of The
Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98). This section states that
reporting is at the facility level with facility defined as any physical property, plant on one
or more contiguous or adjacent properties in actual physical contact Under Subpart W, gas
processing facilities conform to this facility definition with the one minor exception of residue
gas compressors owned and operated by a processing facility, serving only that processing
facility, but located outside the processing facility fence line. Onshore production equipment and
operations are generally located outside the boundaries of a processing facility. Because similar
equipment and operations may be performed at a wellhead in onshore production, such as liquids
separation, compression or dehydration, does not make a wellhead a processing facility anymore
than those same operations and equipment types conducted in a transmission compressor station
or underground natural gas storage site make those facilities processing facilities. In todays
final rule EPA has revised the definition of natural gas processing facilities in Section 98.230, to
include those with fractionation, and those without fractionation with a gas throughput capacity
of 25 mmscfd and above. Please see the preamble Section II.E and Minimum Gas Processing
Throughput in EPA-HQ-OAR-2009-0923. EPA recognizes the possibility of oil and/or gas
producing wellheads being co-located within the boundaries of a processing facility, in which
case those emissions would be reported as part of the processing facility emissions using
1676
methods defined for onshore production equipment. Please see the response to EPA-HQ-OAR
2009-0923-1024-14. In todays final rule EPA has revised the term facility in Section 98.238 for
the purposes of subpart W for onshore production, which groups all equipment under a common
owner/operator, associated with petroleum and natural gas production wells in a basin. Please
see the preamble Section II.D to todays final rule. This definition overrides the contiguous
condition in subpart A of The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR
part 98).
Section 98.236(a): As discussed in the comment provided for the Onshore natural gas processing
plants source category in Section 98.230(a)(3), an onshore natural gas production facility may
contain some of the processes that are currently defined as part of an onshore natural gas
processing plant (e.g., the removal of water, condensate, and/or oil). Due to this confusion in the
definitions of production and processing, many production facilities could fall into both
categories. This will result in different operators defining their operations inconsistently, and will
most likely result in facility emissions being double counted in both source categories. IPAMS
requests that the source categories and definitions for onshore natural gas production facility
and onshore natural gas processing plant clearly define the difference between the operations
to extirpate overlap between the two.
Response: For further details, please see responses to EPA-HQ-OAR-2009-0923-1080-45, EPA
HQ-OAR-2009-0923-1024-14, and EPA-HQ-OAR-2009-0923-1298-24.
Section 98.233(r) Gathering pipeline fugitives
Section 98.232 (c) (9) lists gathering gas pipeline fugitives as a reportable source type for
onshore petroleum and natural gas production, however, 98.233 (r) includes gathering gas
pipeline fugitives only for the natural gas processing sector.
Section 98.233 (r) (3) Onshore natural gas processing facilities shall use the appropriate default
population emission factor listed in Table W-2 of this subpart for fugitive emissions from
gathering pipelines.
As discussed previously, BP recommends designating Onshore Natural Gas Gathering and
Collection as a stand-alone sub-sector of Onshore Petroleum and Gas Production (OPGP)
category, separate from Onshore Petroleum and Natural Gas Production and Onshore Natural
Gas Processing Plants. This provides a clear distinction between production and processing
1677
operations, and eliminates duplicate reporting due to the overlap of some operations and
equipment.
Further, as a separate sub-sector, BP supports EPAs use of the default emission factor for
gathering pipelines, as provided in Table W-1. Natural gas gathering pipelines are widely
dispersed, are generally much smaller diameter, and typically operate at low pressures, resulting
in a lower potential for emissions.
1004-14.
Section 98.230(a)(2) and (3) Definition of the source categories for Onshore Petroleum and
Natural Gas Production and Onshore Natural Gas Processing Plants
As proposed, the definitions under 98.203(a)(2) and (3) lack a clearly defined boundary between
the two segments. While EPA asserts that the facility definition can easily be applied to onshore
natural gas processingsince the operations are all located in a clearly defined boundary
(Technical Support Document p.17), this is not the case in Subpart W, where EPA has expanded
the definition of the natural gas processing plant to encompass additional assets and created an
indistinct boundary as a result. The lack of clarity poses several problems. First, there is
significant overlap between the two definitions in both the physical description and process
descriptions. From the physical description standpoint, both definitions include gathering and
boosting systems and sites which gather gas from multiple wells. By EPAs own
acknowledgement, for gathering pipelines, it is difficult to assign this portion of onshore
production to one particular segment. From the process description standpoint, the definitions
overlap in the areas of phase separation (into hydrocarbon liquids, gas, and water), dehydration
of natural gas, stabilization of hydrocarbon liquids, and removal of NGL if pentane separated as
a component of condensate is considered an NGL. A clear distinction between the production
segment and processing segment must be made to avoid determinations that individual well sites
are processing facilities and to avoid double counting sources under both production and
processing.
Second, structurally, the somewhat simplistic linear construct of the two definitions does not fit
the complexity of the actual industry ownership and operating structure, and may not be
implementable as written. It appears that EPA is assuming gathering lines are dedicated to a
specific gas processing plant. Operatorship/Ownership and routing of gathering and collection
systems between the wells and processing plants is much more complex. Handling of at least
three different patterns of operatorship/ownership must be considered and described with
1678
variations on each definition. These three patterns are:

1. One company operates/owns the wells, collection system, and gas processing facility.
2. One company operates/owns the wells and a second company operates/owns both the
collection systems and processing facilities.
3. One company operates the wells, a second (or perhaps multiple) company operates the
collection systems, and a third company operates the gas processing facilities.
This is further complicated by the fact that many collection systems, in any of the three patterns,
are interconnected or looped to enable balancing of production against available collection,
compression and processing capacity. On any given day a well or collection system booster
station may be routed directly to processing facility A, be routed directly to processing facility B,
or be shuttled to another collection system (or multiple) for routing to either processing facility A
or B or an entirely different processing facility. In each of these scenarios, the operatorship of the
collection systems and processing facilities can be different as described above. Also, in many
instances, certain equipment located on individual well sites (dehydrators are common) are
owned/operated by the collection system operator - which may or may not be the owner/operator
of the processing facility that ultimately receives the gas. As individual gathering/collection
system booster stations are routed to different processing facilities their status can change from
inclusion with the receiving processing facility, inclusion with the wells which they serve, or to
stand-alone facilities (or perhaps not covered due to the stand-alone definition being keyed to
routing directly to a transmission line). This can occur on a day to day basis and make it very
difficult to determine how a particular facility should be handled on a particular day.
EPA could best address the issues described above by modifying the two existing definitions and
including a third definition for Onshore Natural Gas Gathering and Collection Systems, which
segments this portion of the Petroleum and Natural Gas Sector category into three segments
rather than two. The reporting of gathering pipelines/systems as a separate segment is supported
by EPAs Technical Support Document (TSD), which states that Unlike other segments of the
petroleum and natural gas industry, gathering systems may be owned by producers, processing
plants, transmission companies, local distribution companies, or independent gathering
companies. Therefore, it is difficult to assign this portion of onshore production to one particular
segment. One option is to require gathering pipelines to be reported as an emissions source. The
other option is to have a separate segment assigned to gathering pipelines. (TSD p.18).
Following are suggested revisions to clarify the source categories/segments as described above
(proposed language changes/additions are shown in green):
Section 98.230(a)(2) Onshore petroleum and natural gas production. Onshore petroleum and
natural gas production equipment means all structures and equipment associated with wells
(including but not limited to compressors, generators, separation equipment, dehydration
equipment, auxiliary non-transportation related equipment, centralized flow or gathering
stations, central tank batteries or storage facilities), piping (including but not limited to flowlines
or intra-facility gathering lines), and portable non-self-propelled equipment (including but not
1679
limited to well drilling and completion equipment, workover equipment, gravity separation
equipment, auxiliary non-transportation-related equipment, and leased, rented or contracted
equipment) used in the production, extraction, recovery, lifting, stabilization, separation,
compression, pumping, or treating of petroleum and/or natural gas (including condensate). This
also includes associated storage or measurement and all systems engaged in gathering,
separating, treating, compressing, or stabilizing produced gas and hydrocarbon liquids from
multiple wells which is not included in the onshore natural gas gathering and collection systems
or natural gas processing plants , all systems associated with fuel gas use to power onshore
production activities, all systems associated with gas reinjection that is not for underground
natural gas storage purposes, all EOR operations using CO2, and all petroleum and natural gas
production located on islands, artificial islands or structures connected by a causeway to land, an
island, or artificial island which are owned/operated by the owner/operator of the well(s).
NEW Section 98.230 (3)(a)(?) Onshore natural gas gathering and collection systems: Onshore
natural gas gathering and collection systems along with their field gathering and/or boosting
stations are intended to gather and process natural gas from multiple wellsites or other
gathering/collection systems, and compress and transport natural gas (including but not limited
to flowlines or intra-facility gathering lines or compressors) as feed to natural gas processing
plants or directly to a natural gas transmission or distribution facility. Onshore natural gas
gathering and collection systems may include equipment and processes for phase separation,
condensate and water removal, dehydration, stabilization or treating of petroleum and/or
natural gas, compression, storage, and metering. However, the main purpose of an onshore
natural gas gathering and collection system is the gathering and transport of gas prior to
processing or transmission. Onshore natural gas gathering and collection systems do not include
residue gas compression adjacent to and down-stream of a processing facility which are
included with the processing facility if owned/operated by the same owner/operator of the
processing facility
Section 98.230(a)(3) Onshore natural gas processing plants. Natural gas processing plants are
designed to separate and recover natural gas liquids (NGLs) or other non-methane gases and
liquids from a stream of produced natural gas to meet onshore natural gas transmission pipeline
quality specifications through equipment performing one or more of the following processes: oil
and condensate removal, water removal, separation of natural gas liquids, sulfur and carbon
dioxide removal, fractionation of NGLs, or other processes, and also the capture of CO2
separated from natural gas streams for delivery outside the facility. Gas treatment processes
performed on associated gas produced from crude wells for use as fuel gas or for reinjection at
an onshore production facility is not considered a natural gas processing facility for the
purposes of this subpart. In addition, field gathering and/or boosting stations that gather and
process natural gas from multiple wellheads, and compress and transport natural gas (including
but not limited to flowlines or intra-facility gathering lines or compressors) as feed to the natural
gas processing plants are considered a part of the processing plant. Gathering and boosting
stations that send the natural gas to an onshore natural gas transmission compression facility, or
natural gas distribution facility, or to an end user are considered stand alone natural gas
processing facilities. All residue gas compression equipment owned/operated by a processing
plant and located in close proximity to the processing plant, whether inside or outside the
processing plant fence, are considered part of natural gas processing plant.
1680
Response: Todays final rule does not require reporting of emissions from the gathering and
boosting segment of the industry. For more information, see the Preamble, Section II.F to todays
final rule. EPA further clarifies the distinction between onshore production and
gathering/boosting by clarifying that equipment such as compressors, dehydrators, separators and
tanks located on a well pad, or associated with a well pad are onshore production. EPA also does
not agree with excluding gas treatment processes performed on associated gas for use as fuel
gas or for reinjection from the equipment covered by gas processing when that equipment is
within the facility boundaries of a processing facility. Such equipment associated with a well pad
would be onshore production equipment. EPA has also clarified the definitions for onshore
production and onshore gas processing to avoid overlaps; please see response to EPA-HQ-OAR
2009-0923-1080-45 for further details. In todays final rule EPA has revised the definition of
natural gas processing facilities in Section 98.230, to include those with fractionation, and those
without fractionation with a gas throughput capacity of 25 mmscfd and above. Please see
Section II.E of the preamble to todays final rule and Minimum Gas Processing Throughput in
EPA-HQ-OAR-2009-0923.
5. Natural Gas Gathering System Compression Facilities Should Not Be Included In the
Definition of the Term Gas Processing Plant.
Chesapeake supports the comments submitted by Gas Processors Association regarding proposed
Subpart W with respect to the term "gas processing plant" and the need to retain this definition as
understood by industry and as defined under the CAA. See 40 C.F.R. 68.3 . Gas compression
facilities, whether in the gathering systems upstream of gas processing or in the natural gas
transmission sector, are discrete and easily identifiable facilities and should not be included in
the definition of "gas processing plant" under Subpart W.
todays final rule. In the final rule EPA has revised the definition of natural gas processing
facilities in 98.230, to include those with fractionation, and those without fractionation with a gas
throughput capacity of 25 mmscfd and above. Please see the preamble Section II.E and
Minimum Gas Processing Throughput in EPA-HQ-OAR-2009-0923.
Organization: Sierra Club
Commenter: Anne Harvey
With regard to life cycle accounting, EPAs decision to require production and processing
reporting is particularly important because oil and gas companies are increasingly promoting
high carbon products like tar sands oil, liquefied natural gas and oil shale
1681
Response: EPA has retained reporting of emissions from both production and processing
segments of the industry in todays final rule.
Third, API is concerned that the rule includes inconsistencies and overlap in the demarcation of
the different industry segments subject to Subpart W. It would be better if the industry segments
were to be defined along typical ownership and operating arrangements such as producing well
sites and associated equipment, gathering lines in lateral compression, and natural gas
processing. This would avoid duplicative reporting. API suggests that EPA look to the definition
of, quote, transportation in NESHAPs Subpart HHH to define where natural gas production
and processing ends and where natural gas transmission begins.
relates to other segments. EPA has also clarified the boundary between onshore processing and
transmissions compressor stations. For further details, please see Section 98.2 of todays final
rule. EPA has considered the definition for transmission segment from NESHAP HHH and
determined the facility boundary to be similar with Subpart W. The commenter does not provide
details on what specific items in the NESHAP HHH should be considered and how it concerns
Subpart W definition of transmission segment. Hence EPA has retained the definition of onshore
natural gas transmission compression in todays final rule. In the final rule EPA has revised the
definition of natural gas processing facilities in 98.230, to include those with fractionation, and
those without fractionation with a gas throughput capacity of 25 mmscfd and above. Please see
the Section II.E of the preamble to todays final rule and Minimum Gas Processing
Throughput in EPA-HQ-OAR-2009-0923.
14.4
USE OF EQUIPMENT LEAK VERSUS FUGITIVE AND VENTED LEAKS

98.232(b) and page 18643, 98.233(s). Offshore petroleum and natural gas production in both
state and federal waters: These facilities will report emissions from all 'stationary fugitive and
'stationary vented' sources as identified in the MMS GOADS study ... "Greater specificity is
needed to describe' fugitive' and 'vented' emissions. The MMS GOADS study specifically
includes (1) combustion flares, (2) fugitives, (3) glycol dehydrators, (4) loading operations, (5)
flashing losses, (6) mud degassing, (7) pneumatic pumps, (8) pressure/level controllers, (9)
1682
storage tanks, and (10) cold vents. The final rule should indicate whether all of these emission
sources are to be included in the emissions report.
Response: In todays final rule, the definition for equipment leaks has replaced the definition
for fugitive emissions, however the definition for vented emissions has not changed. The rules
definition for equipment leaks is consistent with the use of the term in the Alternative Work
Practice to Detect Leaks from Equipment for 40 CFR parts 60, 63, and 65. Todays rule has also
revised the monitoring requirements for offshore production to be consistent with MMS GOADS
and therefore equipment leaks, vented and flare sources included in GOADS are subject to
reporting under subpart W.

Pre-amble, page 30, The term fugitive emissions has come to be interpreted differently in
various jurisdictions around the world.
- CAPP recommends the use of fugitive equipment leaks as it is more descriptive of the
emission source being targeted.
Response: Please see the response to comment EPA-HQ-OAR-2009-0923-0847-6.

Clarifications or Definitions are Required for Component Types and Related Survey Criteria
Vented and fugitive emissions are better defined in the Proposed Rule than in the 2009 version of
Subpart W. However, consistent nomenclature is ultimately required to eliminate ambiguity,
redundancy (i.e., double counting), and unnecessary confusion in rule implementation. For
example, compressor blowdown valve is listed as a component in Table W-3 and the
associated EF is more than an order of magnitude larger than any other listed. It appears that the
emissions are not related to a leaking valve body or stem, but rather natural gas bypassing the
valve seat, through the valve, and vented out a blowdown vent stack. Thus, the survey point
for leak detection is the blowdown vent even though the source is a fugitive leak through an
unseated valve. The Proposed Rule presumes a certain level of expertise and understanding of
facility-level components, operation, and similar issues, and there is no ability for the reader to
discern this point. Similar ambiguity in the Final Rule could cause significant implementation
and compliance issues.
The lack of clarity regarding component names, component types, associated leak source, and
leak survey criteria for the emission point will result in misapplication of emission factors,
1683
survey errors, and reporting errors. INGAA recommends that clarity and definition of the source
and measurement method(s) be addressed through the EF reference document discussed above.
There must be clear, concise definitions and explanations within Subpart W, especially regarding
differentiation among the sources and their associated emission estimation criteria.
INGAA recommends that additional clarity be added to the Final Rule. This may best be
accomplished by including component definitions, source descriptions, and related leak survey
criteria with citation to the EPA technical reference document discussed above. Although
INGAA strongly advocates this separate document for emission factors, it is imperative that the
document be available concurrent with the Final Rule because it will play an important role in
compliance and implementation. INGAA offers its assistance in development and peer review of
this important technical resource.
Response: The commenter has misinterpreted the intent of the rule. The emission factors in
Table W-3 in the rule are for equipment leaks from components, not for venting emissions.
Definitions of sources and components are located in either subpart W, or in the overarching
MRR. As the placement of emissions factors in a separate reference document would not
decrease the time necessary to update subpart W specific factors and references, EPA will not
remove the emission factors from subpart W itself.
Commenter:
EPA seeks comments on the use of the term equipment leak versus fugitive and vented as
defined in the proposed supplemental rule. (page 30)
Equipment leak and fugitive can be used interchangeably. A vented emission should be
characterized as intentional or unintentional operational release. The key difference between
fugitive and vented is that fugitive emissions are unknown releases that must be calculated
either by direct measurement or analytical methods; whereas vented emissions are generally
designed into the operation of the system.
Preamble p. 30 EPA seeks comment on the use of the term equipment leak versus fugitive
and vented as defined in the proposed supplemental rule.
API supports the use of terms fugitive and vented to describe the non-combustion emission
sources associated with oil and natural gas systems. (Please refer to earlier definition of fugitive
1684
emissions provided in Section III(1) of this document.) Although the term equipment leak is
consistent with fugitive, it is incorrect to apply the term to vented emission sources.
EP A should define "fugitive" emissions consistent with other Clean Air Act ("CAA") regulatory
programs.

Definition of Fugitive Emissions. "EPA seeks comment on the use of the term 'equipment leak'
versus 'fugitive ' and 'vented' as defined in the proposed supplemental rule. " (75 FR 18613)
Resolute Comments:
EPA should define fugitive emissions in Proposed Subpart W consistent with existing rules
promulgated under the CAA to avoid confusion among various regulatory programs. EPA should
apply the definition of fugitive emissions already found in 40 CFR 52.21(i)(20) and 40 CFR
63.2, which state that "fugitive emissions means those emissions that could not reasonably pass
through a stack, chimney, vent or other functionally equivalent opening." For example, leaks
from natural gas driven pneumatic controller and valves, pump seals, non-pneumatic pumps and
connectors should all be considered fugitive emissions. The use of an additional term, such as
"equipment leak," will only introduce additional confusion and increase the reporting burden
without providing additional useful information to the EPA.
On page 43 of the Technical Support Document and in 98.233(p)(4), the EPA addresses fugitive
emissions from compressor operations at different operational modes. In the Technical Support
Document, the EPA concedes that to address this issue, operators must measure emissions for
each mode the compressor is operated in and the duration of that operational mode and this will
increase the reporting burden, since measurements will have to be taken at each mode of
1685
compressor operation. Indeed this requirement will greatly increase the reporting burden
while not addressing a significant source of GHG emissions. 98.233(p)(4) requires measurement
in several operational modes: Compressor engines are generally either operational or not. Some
of the operational modes are not normally encountered and this would have to be artificially
reproduced in the field for the sole purpose of reporting. A compressor would not standby
pressurized for any amount of time that would affect its emissions significantly as it would
otherwise be offset by the fact that the compressor is operational. Therefore, requiring
measurement in each mode is burdensome, difficult to schedule, and does not reflect a significant
source of emissions.
Response: EPA does not agree with this comment. Typically, reciprocating compressor rod
packing emissions vary with the mode of operation of the compressor. The emissions are highest
when the compressor is operating and lower when they are in standby pressurized mode. Rod
packing seals become worn with time and then will typically leak more. Note that rod packing
wear is highly variable with all compressors, even those that are the same make and model.
When the compressor is not operating depressurized, there may be leakage of natural gas through
the unit isolation valve, particularly if the valve seat has become fouled and will not completely
close. Hence to correctly characterize annual emissions from reciprocating and centrifugal
compressors, estimation of emissions in three compressor modes, operating, standby pressurized,
and not operating depressurized is required. A temporary meter such as vane anemometer or
permanent meter such as orifice meter can be used to measure emissions from the vents.
However, EPA agrees to allow compressor venting measurement in the as found mode, see
The ambiguity over the definitions of the several items in the proposed rule, availability of
sufficient numbers of "optical imaging instruments" for leak detection of fugitive emissions,
availability of trained personnel- both contractors and facility personnel- to operate the "optical
imaging instrument" makes compliance with the proposed rule nearly impossible to achieve.
Response: In regard to definitions, please see the response to comment EPA-HQ-OAR-2009
0923-1039-19. In regard to difficulty with complying with the rule, EPA is allowing Best
Available Monitoring Methods for certain sources and time periods, please see Section II.F of
14.5
BOUNDARY ISSUES WITH SUBPART RR

1686

(Preamble p. 85) Generally, EPA has proposed that onshore production be reported at the basin
level, as opposed to the unit or field level, to minimize reporting burden. EPA notes that in a
concurrent proposed rulemaking for facilities that conduct CO2 injection or geologic
sequestration (subpart RR), the term facility is defined at a more disaggregated level,
specifically as a well or group of wells. EPA seeks comment on the use of more disaggregated
reporting options for subpart W.
As noted in the comments to questions W3 through W5, the API does not agree that the basinlevel facility definition minimizes the reporting burden. That said, the facility definition should
be consistent for Subparts W and RR, as CO2 injection is a technique used to enhance oil
production, and thus some entities reporting under Subpart RR will also be reporting under
Subpart W.
Response: EPA does not agree with the commenter. The proposed subparts RR and W have very
different GHG sources that need to be characterized, and the facility and source category
definitions have been chosen accordingly. The proposed subpart RR requires reporting from
facilities that inject CO2 underground for geologic sequestration. The proposed subpart RR uses
the definition of facility in 40 CFR 98.6 and are focused on reporting quantities of CO2 received
and sequestered. On the other hand, Subpart W is primarily concerned with above surface
equipment associated with the wellhead in addition to the wellhead itself. EPA considered
requiring the proposed subpart RR reporting at the basin level, but concluded that data on CO2
received and sequestered is most useful to EPA at a project level. This is because the proposed
subpart RR applies to the injection of CO2 in a variety of geologic formations beyond oil and gas
fields, and basin level reporting may not be applicable in all geologic circumstances.
An EOR site permitted as Class II under the Underground Injection Control (UIC) program is
required to report information on the quantities of CO2 received and the source of CO2. An EOR
site is not required to report under the proposed subpart RR unless it chooses to opt-in to the
proposed subpart RR or is permitted as Class VI under the UIC program. An EOR project
reporting under the proposed subpart RR must calculate equipment leak and vented CO2
emissions from surface equipment between the flow meters and the wellhead using calculation
procedures provided in subpart W. EPA concluded such data are important in order to provide a
proper accounting of the amount of CO2 that is geologically sequestered.
Hence, EPA has chosen the facility definitions for proposed subpart RR, and W and has retained
them in todays final rule.
14.6
OTHER

Page 18614, 2nd column, 3rd paragraph: The sentence "Extraction includes several types of
processes: reservoir management, primary recovery, secondary recovery such as down-hole
1687
pumps, water flood ... " is not quite correct as the use of down-hole pumps is not a secondary
recovery process, but rather a form of artificial lift. Secondary recovery processes work with the
forces in the reservoir by using methods such as water flooding and some gas injection
technologies. We suggest changing the sentence to "Extraction includes several types of
processes: reservoir management, primary recovery, artificial lift, such as down-hole pumps,
secondary recovery such as water flood ..."
Page 18615, 3rd column, 1st paragraph: It is stated that offshore petroleum and natural gas
production CO2 and CH4 emissions accounted for 5.1 million metric tons CO2e. In Table W-2 on
page 18618 the total emissions from offshore petroleum and gas production are 11.3 million
metric tons CO2e. Please ex plain the differences. Does the first figure not include combustion?
Response: EPA agrees that clarifying its description of extraction to not include downhole
pumps as secondary recovery would be useful to reporters. Todays final rule describes
extraction to include several types of processes, reservoir management, primary recovery,
artificial lift, secondary recovery such as water flood
With regard to the emissions cited in the Preamble, EPA clarifies in todays final rule that there
were a total of 11.3 million metric tons CO2e emitted in 2006 by offshore petroleum and natural
gas production, as was cited in Table W-2, with 5.1 million metric tons CO2e predicted
emissions coverage under subpart W at the 25,000 metric ton threshold.

Page 18633, 3rd column, 98.6 Definitions. Acid Gas: The definition should read " ... the
contaminants that are separated from sour natural gas by an acid gas removal unit."
Page 18634, 3rd column, Flare Combustion: The definition reads "Flare combustion means
unburned hydrocarbons including CH4, CO2, N2O ..." This is inaccurate as most of the emissions
from the flare are combustion products. Since a flare cannot achieve 100% burn efficiency, a
small percentage of the hydrocarbons (including CH4) leave the stack without being combusted.
The definition should be modified accordingly.
Response: EPA agrees that there was a grammatical error in the definition of acid gas. For
more information see response to EPA-HQ-OAR-2009-0923-1151-17.
With regard to flare combustion, EPA agrees that the definition of flare combustion emissions
was confusing. In todays final rule the term has been modified to flare stack emissions and
clarifies that it means CO2 and N2O from partial combustion of hydrocarbon gas sent to a flare
plus CH4 emissions resulting from the incomplete combustion of hydrocarbon gas in flares.
1688

Flare combustion definition: The Proposed Rule includes additional and revised definitions in
98.6. The definition of flare combustion is poorly worded because CO2 and N2O are not
unburned hydrocarbons or products of incomplete combustion. INGAA recommends the
following revision: Flare combustion means emissions from combustion of gas in flares
unburned hydrocarbons including CH4, CO2, and N2O emissions resulting from the incomplete
combustion of gas in flares.
Response: EPA agrees, and has revised the definition of flare emissions in todays final rule.

Proposed Rule Section 98.6 Definition for Flare Combustion. A revision to the proposed
definition is requested for clarity and accuracy.
Response: EPA agrees, and has revised the definition of flare combustion in todays final rule.

Flare combustion
The definition of flare combustion is poorly worded. We suggest the following:
Flare combustion emissions means emissions of CO2 and N2O formed during the combustion
process and emissions of CH4 resulting from the incomplete combustion of gas in flares.
Onshore petroleum and natural gas production owner or operator
The proposed definition on page 18635 has an incorrect reference to 98.230(b)(2). The correct
reference is 98.230(a)(2).
1689
Response: While EPA does not agree with the commenters suggested text, EPA has revised
the definition of flare emissions; please see the response to EPA-HQ-OAR-2009-0923-0847-4.
With regard to the definition of onshore petroleum and natural gas production owner or operator,
EPA agrees that it referenced the wrong paragraph, as discussed in this comment. Todays final
rule references the appropriate paragraph that defines the onshore petroleum and natural gas
production sector in the definition.

Acid Gas CAPP requests that the EPA modify the definition to: means hydrogen sulfide
(H2S) and/or carbon dioxide (CO2) contaminants.. This change would be consistent with the
definition currently used by the sector.
Condensate CAPP requests the EPA implement the following change to the definition of
condensate: means hydrocarbon and other liquid separated from natural gas that condenses due
to changes in the temperature, pressure or both and remain liquid at storage conditions
atmospheric conditions..
Flare Combustion Efficiency - CAPP requests the EPA implement the following change to the
definition of Flare combustion efficiency: means the fraction of natural process gas, on a
volume or mole basis
Response: EPA agrees, and has revised the definition of acid gas in todays final rule. Please
With regard to the definition of condensate, EPA disagrees that the proposed definition further
clarifies what subpart W means by the term condensate. The primary clarification sought by
the commenter was that storage conditions are atmospheric. EPA disagrees that this clarification
is necessary, pressurized storage of condensate is a measure to avoid emissions. Since this is a
greenhouse gas emissions reporting rule, EPA only expects that emissions will be vented from
atmospheric storage of condensate, thus the proposed clarification would not affect the reporting
under subpart W.
With regard to the definition of flare combustion efficiency, EPA agrees that it will be less
confusing to reporters to clarify the use of the term natural gas in the definition. In todays
final rule, 98.238, the definition of flare combustion efficiency is the fraction of hydrocarbon
gas, on a volume or mole basis, that is combusted at the flare burner tip.
1690

In 98.236(c)(14)(iv), the EPA requires facilities to identify the proportion of total natural gas to
pure hydrocarbon stream being sent to the flare annually for the reporting period. CAPP requests
clarification. Is the EPA interested in the quantity of purge gas required to ensure a continuous
flame? Or is the request intended to require quantification of non-hydrocarbon components of
the natural gas stream. Additionally CAPP requests a definition of total natural gas and pure
hydrocarbon stream.
Response: Upon further analysis and review, EPA has determined in todays final rule that the
reporters can use engineering estimates based on process knowledge and best available data and
operating records to determine the amount and type of gas sent to a flare on an annual basis. This
would encompass all gas sent to the flare, including purge and pilot gas as well as waste gas. The
terms total natural gas and pure hydrocarbon stream are no longer used in today-s final rule.

Clarification of "Large Compressors"
EPA refers to "large compressors" in multiple places in the proposed rule. Red Cedar is
requesting a definition or clarification of the term "large compressors" as it relates to this rule.
Red Cedar recommends a horsepower rating of 2,000 hp and greater for the compressor driver be
used to define a "large compressor".
Response: Todays final rule no longer uses the term large compressor. This term was used
in the context of onshore petroleum and natural gas production compressors to differentiate
between wellhead compressors, which are typically smaller and widely dispersed, and gathering
compressors which range widely in size and are often in groups of multiple compressors.
Onshore natural gas processing, onshore natural gas transmission compression, underground
natural gas storage, LNG storage, and LNG import and export equipment compressors are much
larger on average than compressors in the onshore petroleum and natural gas production
segment. EPA determined that it is too burdensome for the contribution of emissions that are
estimated to be released from these smaller production compressors to perform leak detection
and measurement. Therefore, todays final rule requires all onshore production compressors to
estimate emissions using a factor. Todays final rule also does not include gathering lines and
boosting stations. For further clarification, please see Section II.F of the preamble. For other
industry segments where multiple compressors are generally grouped in a facility EPA
determined that leak detection and vent measurements could be accomplished cost-effectively,
provided reporters spread the emissions measurements out over a three year period. For more
1691
information on selection of sources to report, please see chapter 4 of the background technical
support document of the rule making docket, EPA-HQ-OAR-2009-0923).

Commenter: Kim Dang
Comments on Definitions. In addition to our comments on the definitions of facility types within
the petroleum and natural gas sector (discussed above in section II.B.2), we also offer the
following comments on other definitions in the proposed text of Subpart W:
1. Fugitive emission. EPAs proposed definition of the term fugitive emission 254 appears to be
drawn from the agencys existing PSD regulations. 255 The stipulation in this definition that
fugitive emissions only include those emissions that could not reasonably pass through a stack,
chimney, vent, or other functionally equivalent opening is inappropriate for a rule that is
focused on reporting, rather than control of emissions. In the PSD context, fugitive emissions are
defined with reference to stacks, chimneys, and vents in order to require sources to capture and
contain unintentional emissions wherever reasonably possible. The Mandatory Reporting Rule
does not have this objective, and should not require sources in the petroleum and natural gas
sector to make a judgment call as to whether a particular source could be channeled through a
vent or other opening. Therefore, Kinder Morgan recommends that EPA modify the proposed
definition of fugitive emissions to more closely reflect the usage of that term in our industry. A
recommended definition appears in Appendix A of these comments.
2. Flare combustion. Kinder Morgan recommends that EPA reconsider the proposed definition of
flare combustion, 256 which currently includes unburned hydrocarbons including CH4, CO2,
and N2O. This definition is problematic because the petroleum and natural gas industry does not
use the term flare combustion to refer to uncombusted hydrocarbons and combustion
byproducts. Rather, flare combustion has a well-established definition in the industry, referring
to the process of combusting gases in a flare device. EPA should either establish a separate term
for unburned hydrocarbons or simply refer to these substances directly in the text of the rule.
3. Blowdown vent stack emissions. The proposed definition of the term blowdown vent stack
emissions 257 refers only to releases of natural gas, even though Subpart W calls for emissions
monitoring of CO2 injection equipment that does not handle natural gas. In addition, the
proposed definition includes gas released due to maintenance, a description that is overly
broad and could capture emissions that are not traditionally regarded as blowdown emissions. To
address these issues, Kinder Morgan recommends that the definition, and corresponding
254
Proposed Subpart W, 75 Fed. Reg. at 18,634.] appears to be drawn from the agencys existing PSD regulations.
255
See 40 C.F.R. SECTION 52.21(b)(20).
256
Proposed Subpart W, 75 Fed. Reg. 18,634.
257
Proposed amendments to 40 C.F.R. SECTION 98.6.

1692
quantification methodology, be clarified to include equipment in CO2 service as well as

equipment handling natural gas. In addition, the definition should be revised to include gas
trapped within a compressor and associated piping that is released during all start-up and shut
down procedures for that compressor, as opposed to gas released during all maintenance
operations.
4. Natural gas high- and low-bleed pneumatic devices. The proposed methodologies for
quantifying emissions from high and low-bleed pneumatic devices powered by pressurized
natural gas refer to continuous bleed devices,258 yet the definitions of high- and low-bleed
pneumatic devices do not. 259 To avoid confusion, Kinder Morgan requests that EPA clarify the
definitions of high- and low-bleed pneumatic devices to refer only to devices with a continuous
bleed function, as opposed to intermittent bleed function.
5. Definition of Transmission Pipelines. EPA should omit the adjective crosscountry in its
definition of transmission pipelines, since not all transmission pipelines subject to Subpart W
cross the country or even cross state lines.
Response: EPA disagrees that it must define the term fugitive emission so that it does not
stipulate that the emissions could not reasonably pass through a stack, chimney, vent, or other
functionally-equivalent opening. One practical example from subpart W is that through-valve
leakage from compressor unit and blowdown valves may be manifolded into a vent stack and are
covered under a different methodology than equipment leaks and emission factors. EPA does
not anticipate that reporters will be unable to determine what emissions are considered to be
equipment leaks by this definition and it serves a purpose, as stated, under subpart W. In
addition, EPA has replaced the term fugitive emissions with equipment leaks. This change
was made to ensure consistency with the terminology in the Alternative Work Practice to Detect
Leaks from Equipment for 40 CFR parts 60, 63, and 65.
EPA agrees, and has revised the definition of flare combustion in todays final rule. Please see
the response to EPA-HQ-OAR-2009-0923-0847-4.
EPA agrees that it would be less confusing to reporters to clarify the definition of blowdown vent
stacks to include CO2 as well as natural gas. Todays final rule applies to all blowdown events,
not only maintenance shutdowns, and specifies calculating the volume of all equipment, piping
and vessels between isolation valves adjusted to standard cubic feet based on actual pressure and
temperature before blowdown. Also, blowdown events that are less than 50 standard cubic feet
need not be reported to reduce burden and uncertainty in small blowdowns such as level sight
glasses, instrument leads, etc.
EPA has considered commenters suggestions regarding characterization of pneumatic devices
and has agreed to prescribe the use of emission factors provided in GRI and the API
Compendium. As a result, todays final rule requires counting pneumatic devices that are
258
See Proposed 40 C.F.R. SECTION 98.233(a)(1)(ii), (b).
259

1693
continuous high-bleed, continuous low-bleed, and intermittent bleed, and applying emission
factors for each type.
EPA agrees that it is less confusing to reporters to clarify the definition of transmission pipelines
to include those that are not cross country. Todays final rule includes both interstate and
intrastate pipelines in the definition.

98.6 Definitions: There is no definition for Offshore petroleum and natural gas production
owner or operator, as there is for Onshore petroleum and natural gas owner or operator.
OOC Comment: See comment re: 98.3(c)(11) above. Due to the complexities of offshore facility
ownership and operatorship, we feel a definition is warranted. A facility can have multiple
owners and operators, and the number of fields and wells with multiple owners that tie back to
the facility can further complicate who reports what to the EPA. We feel the EPA should define
the offshore facility owner/operators in a manner similar to that of the onshore definition, where
facilities with multiple owners/operators are instructed to designate one entity to report all
emissions from the joint facility, and where production handling from fields not apply to the
definition of facility ownership.
Response: EPA disagrees that it has not adequately defined who must report for offshore
petroleum and natural gas production facilities covered under subpart W. Todays final rule
instructs reporters to use the methodology set forth by the most recent Gulfwide Emission
Inventory Study (often referred to by its activity collection portion, GOADS, by industry). In
following this methodology, owners and operators will determine who reports the same as they
would under the MMS regulation. For non-GOADS reporting platforms, including state waters
and state and federal waters outside the Gulf of Mexico, please refer to Section V of the
preamble to The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98), and
Section II.F of the preamble to todays final rule, for the designated representative who must
certify the emissions inventory report. The designated representative (DR) is the entity that is
responsible for submitting the emissions data pursuant to todays final rule. Please see the

98.6 Definitions: Acid gas removal vent stack emissions mean the acid gas separated from the
acid gas absorbing medium () and released with methane and other light hydrocarbons to the
1694
atmosphere or a flare.
OOC Comment: There should be a separate definition for Acid gas removal flare stack, as
defining vent stack emissions as including flare emissions may create confusion.
98.6 Definitions: Dehydrator vent stack emissions means natural gas released from a natural gas
dehydrator system
OOC Comment: Several dehydrator systems release the natural gas back into the process rather
than directly to the atmosphere. There should be mention that the natural gas is released to the
atmosphere.
Response: EPA disagrees that it should define acid gas removal flare stacks separately from
vent stacks. Whether the AGR stack leads directly to the atmosphere or is routed to a flare
system does not affect the methodology to determine the volume of CO2 only emitted to the
atmosphere. Reporters must use this methodology to determine the volume passing through the
AGR stack, then if it is vented to the atmosphere, that value is reported as an emission, if it is
passed to a flare system, the methodology accounts for the CO2 that passes unchanged through
the flare to the atmosphere, not counting any hydrocarbon component of the AGR vent nor other
hydrocarbons sent to the flare.
EPA disagrees that it must caveat the term dehydrator vent stack to only mean natural gas that is
released to the atmosphere. The commenter suggests that for some natural gas dehydration
systems, the gas that would be vented is captured and injected back into the process. Since this
is not a release of gas, EPA does not deem it necessary to caveat the definition.

Section 98.230(a)(1): Offshore petroleum and natural gas production. Offshore petroleum and
natural gas production is any platform structure, affixed temporarily or permanently to offshore
submerged lands, that houses equipment to extract hydrocarbons from the ocean or lake floor
and that transfers such hydrocarbons to storage, transport vessels, or onshore. In addition,
offshore production includes secondary platform structures and storage tanks associated with the
platform structure.
OOC Comment: The rule should clearly state that it does not apply to offshore drilling facilities
unless they are included by GOADS. The category description does not include facilities
engaged only in drilling (for example, Mobile Offshore Drilling Units or MODUs) since, in
contrast to production platforms, their primary purpose is not extracting and transferring
hydrocarbons. Similarly, GOADS 2008 guidance only included drilling activities in cases of
platform rig applications or jack-up rigs associated with an Area/Block with existing production
structures. Thus, the rule should clearly state that drilling activities are not included or point to
1695
GOADS to determine which ones are. If MODUs were to be included, it should be the MODU
owner/operator who does the reporting.
Response: EPA agrees, and has clarified todays final rule. Offshore drilling and exploration
conducted on a production platform are included in subpart W. Subpart W does not include
reporting of emissions from offshore drilling and exploration that is not conducted on production
platforms.

The: Condensate definition states that it includes both water and hydrocarbon liquids. Although a
condensate storage tank in a production field typically includes both water and condensate, the
IWO separate into different phases and are removed from the tank as separate streams. The tank
operator documents the condensate production volume as the condensate phase that is removed
from the tank. It is not accurate to state that condensate includes water, and we request that EPA
remove "water" from the "condensate" definition.
Response: EPA disagrees that it should remove water from the condensate definition. EPAs
intent is to collect emissions data from storage tanks immediately downstream of multi-phase
separators such as pressurized gas-liquids separators at well production sites and scrubbers at the
inlet of compressor stations. These are targeted because previous studies have identified that the
separator dump valves in some cases malfunction, allowing gas to vent through the storage tank,
and in the case of production separators, significant volumes of gas flash out of solution with the
hydrocarbon phase in the atmospheric tank. Because the liquids collected in storage tanks from
these separators may contain substantial volumes of water, and particularly the case in tanks
collecting condensed water from transmission scrubbers, EPA specifically included water in the
definition of condensate to capture the emission sources.

The Conventional Wells definition stales that they do not employ hydraulic fracturing.
"Conventional" gas reservoirs are typically underground formations composed of sandstone,
limestone, dolomite, and/or mixed mineralogy, and "conventional" wells are those that produce
from these types of reservoirs. The term "conventional" does not necessarily consider whether
the formation has been hydraulically fractured. Many of the "conventional"' and
"unconventional" wells in the United States have been hydraulically fractured. We request that
EPA define the terms "conventional wells" and "unconventional wells" based on the well's
reservoir type, as opposed to whether the well operates in a fractured reservoir.
1696
Alternatively, the rule could omit the terms "conventional" and "unconventional" and replace
these terms where needed with a reference to fracturing. So, for example, the emission
calculation methodology section could be retitled from Gas well venting during unconventional
well completions and workovers" to ''Gas well venting for hydraulically fractured well
completions and workovers."
Response: EPA agrees that its characterization of conventional and unconventional well
completions and workovers did not directly address the reasons for the separate monitoring
methods. Todays final rule no longer uses the language of conventional or unconventional
wells, but instead refers to well completions and workovers with hydraulic fracture and those
without hydraulic fracture.

The Unconventional Wells definition states that they are wells that employ hydraulic fracturing.
The term ''unconventional'' applies to wells that produce from reservoirs that are not
"conventional," most often coal beds and shales. In addition, the term ''unconventional typically
refers to wells that produce from tight formations with a permeability less than 0.1 milli-Darcies.
The term "unconventional" does not necessarily consider whether a reservoir has been
hydraulically fractured. Many of the "conventional-' and '"unconventional" wells in the United
States have been fractured. We request that EPA define the terms "conventional wells" and
unconventional wells" based on the well's reservoir type, as opposed to whether the well
operates in a fractured reservoir.
Response: EPA does not agree with this comment. The terms conventional and
unconventional are no longer used in todays final rule, replacing unconventional with
hydraulic fractured. Please see the response to EPA-HQ-OAR-2009-0923-1040-13.

The Gas to Oil Ratio (GOR) definition states that it is the ratio of the volume of gas at standard
temperature and pressure produced from a volume of oil when depressurized to standard
temperature and pressure. The oil and gas industry does not typically measure GOR in this
manner. GOR is usually expressed as the measured natural gas volume that separates at actual
separator temperature and pressure (corrected to standard temperature and pressure) divided by
the measured oil volume from the storage tank(s). We request that EPA modify the GOR
definition to accurately reflect the current industry-accepted method.
1697
Response: EPA disagrees with this comment, but has revised todays final rule methods in
98.233(l) and 98.233(m), such that reporters can use existing data, an appropriate standard
method published by a consensus-based standards organization, or an industry standard practice
to determine GOR. Thus, the method of determining GOR will coincide with industrys typical
measurement. However, for the purposes of reporting to subpart W, all emissions must be
converted to standard conditions. Thus, reporters must convert the measured GOR to standard
volume per liquid barrel or convert actual emissions estimated using GOR to standard conditions
for reporting (both are mathematically equivalent).

The Reservoir definition states that it is characterized by a single natural gas pressure system.
This statement is not accurate because a reservoir can have multiple pressure systems that are
separate pods with permeability barriers between them. We request that EPA remove the last
sentence from this definition.
Response: EPA agrees, and todays final rule deletes the reference to a single natural gas
pressure system. This term reservoir is not used in any calculation methodology or reporting
requirement in todays final rule. This term is simply a descriptor used, for example, in
describing underground gas storage in a depleted gas or oil reservoir or enhanced oil recovery
in a crude oil reservoir.

The Well Completions definition states that it is a process that allows for the flow of petroleum
or natural gas from newly drilled wells to expel drilling and reservoir fluids and test the reservoir
flow characteristics. This definition is not accurate. The term well completion" typically refers
to the process of connecting the well bore to the reservoir. It may include treating the formation
or installing tubing, packer(s), or lifting equipment. We request that EPA modify the "well
completions" definition to reflect the description given here.
Response: EPA disagrees that the definition of well completion is not accurate. EPA
understands that installing tubing, packer(s), or lifting equipment may also accompany
connecting the well bore to the reservoir; however, a well is not completed until the well bore is
connected to the reservoir, so subpart W will not miss any completions. In addition, if lifting
equipment is installed or maintenance is performed on a well already producing, then that is
considered a workover for the purposes of this rule.
1698

The Well Workover definition states: "This process also includes high-rate back-flow of injected
water and sand used to re-fracture and prop open new fractures in existing low permeability gas
reservoirs." We request that EPA modify this sentence to read, "This process may also include
high-rate flowback of injected gas, water, oil and sand used to fracture or refracture and prop
open new fractures in existing lower permeability gas reservoirs."
Response: In todays final rule, EPA incorporated the language this process also includes
high-rate flowback of injected gas, water, oil and proppant used to fracture or re-fracture and
prop-open new fractures in existing low permeability gas reservoirs

EPA proposed definition: Enhanced Oil Recovery (EOR) means the use of certain methods such
as water flooding or gas injection into existing wells to increase the recovery of crude oil from a
reservoir. In the context of this rule, EOR applies to injection of critical phase carbon dioxide
into a crude oil reservoir to enhance the recovery of oil.
Modify the definition as follows: Enhanced Oil Recovery (EOR) means the use of certain
methods such as water flooding or gas injection into existing wells to increase the recovery of
crude oil from a reservoir. In the context of this rule, EOR applies to injection of recovered
carbon dioxide into a crude oil reservoir to enhance the recovery of oil.
The purpose of this requested change is to acknowledge that some oil production fields re-inject,
into the reservoir, the natural gas that is produced with the oil. This natural gas has some small
quantity of naturally occurring CO2 in it and is not injected for the purposes of enhanced oil
recovery.
Response: EPA disagrees that the suggested revision is necessary to clarify the definition of
enhanced oil recovery subject to subpart W. EPA is requiring reporting of only the critical phase
CO2 operations associated with EOR, and recognizes that there are other operations designated
as enhanced oil recovery which are not subject to todays final rule.
1699

Flare combustion
Title of this definition should be expanded to Flare combustion emissions as it more accurately
represents the gasses represented in the subpart.
Response: EPA agrees, and, in todays final rule, EPA has revised the term flare combustion.
Todays final rule uses the term flare stack emissions, please see the response to EPA-HQ-OAR
2009-0923-0847-4.

High-Bleed Pneumatic Devices
Pneumatic devices in the oil and gas field are also powered by compressed air. This definition
should specify that devices powered by natural gas or gas containing reportable compounds are
included in this rule or change the defined word to natural gas driven high-bleed pneumatic
devices.
Response: EPA disagrees that it is necessary to change the term high-bleed pneumatic device to
natural gas driven high-bleed pneumatic device. The definition states for subpart W that the
covered devices are powered by pressurized natural gas and the methodology in todays final
rule states that it is for natural gas pneumatic devices. EPA does not anticipate that any reporter
will confuse the source or monitoring methods to apply to air-driven devices.

Low-Bleed Pneumatic Devices
Pneumatic devices in the oil and gas field are also powered by compressed air. This definition
should specify that devices powered by natural gas or gas containing reportable compounds are
included in this rule or change the defined word to natural gas driven high-bleed pneumatic
devices.
Response: EPA disagrees that it is necessary to change the term low-bleed pneumatic device to
natural gas driven low-bleed pneumatic device. Please see the response to EPA-HQ-OAR-2009
0923-1058-6.
1700

Commenter:
General Comments Clarification of Definitions
Yates believes the following terms require clarification or definition in the rule:
98.232(c)(9): Gathering pipeline fugitives. Yates Petroleum Corporation believes that it is
unclear when Production is required to include gathering, and when boosting/processing includes
gathering. Yates Petroleum Corporation reiterates that it is unclear in the rule where Production
site responsibility ends and Gathering/Processing site responsibility begins.
98.232(i)(2): Below ground meter regulators and vault fugitives. It is unclear in the supporting
documentation why EPA is requiring below-ground fugitives.
98.232(j): Flares. Yates Petroleum Corporation believes that the EPA needs to clarify the
definition of this emission source. Are enclosed flares or fireboxes included in this source
category?
98.233(a)(1)(ii): Calculate the natural gas emissions for each continuous bleed device using
Equation W-1 of this section. EPA should define continuous.
98.233(g)(ii)(C): Gas well venting during unconventional well completions and workovers. The
term producing field in this subpart is not defined.
98.230(a)(3): Onshore Natural Gas Processing Plant Source Definition. In NSPS KKK, a natural
gas processing plant is defined as, any processing site engaged in the extraction of natural gas
liquids from field gas, fractionation of mixed natural gas liquids to natural gas products, or both.
Nonfractionating plant means any gas plant that does not fractionate mixed natural gas liquids
into natural gas products. EPAs treatment of onshore natural gas processing plants does not
appear to align with the NSPS KKK definition of natural gas facility. For purposes of
applicability, a gathering/boosting station without gas processing is not subject to the
requirements of KKK. Therefore, a gathering/boosting station that is not owned by a processing
plant should not be considered part of that processing plant facility.
Response: With regard to gathering pipeline equipment leaks, EPA is not requiring the
monitoring and reporting of emissions from gathering and boosting equipment at this time.
Please see the discussion in Section II.F of the preamble to todays final rule for more
information.
Regarding the process it used to determine a methodology for below ground metering and
regulator stations, see detailed discussion of its general methodology and reasoning for
1701
including this source in chapter 4.c of the background technical support document in the rule
making docket (EPA-HQ-OAR-2009-0923).
EPA has reviewed this comment and notes that the flare source in todays final rule does not
include enclosed flares (e.g. incinerators) or the firebox of enclosed thermal oxidation units. For
further details, please see the response to EPA-HQ-OAR-2009-0923-1015-31.
EPA agrees with the comment. For a clarification on the meaning of continuous, please see the
EPA has defined the term field in the rule and a producing field simply means a field that has
any production in the reporting period.

Certain definitions in the proposed rule should be amended. Proposed 98.6
defines "flare combustion" to mean ''unburned hydrocarbons including CH4, CO2, N2O emissions
resulting from the incomplete combustion of gas in flares." This definition is somewhat peculiar
in that it includes CO2 and N2O which in fact are products of combustion, rather than the result
of non-combustion. Hence, TPA asks EPA to review this list and consider eliminating CO2 and
N2O.
In addition, TPA suggests that the definition of "flare combustion efficiency" be amended as
follows:
''the hydrocarbon fraction of [strike through: natural gas] waste stream, on a volume or mole
basis, that is combusted at the flare burner tip."
Response: EPA agrees, and has revised the definition of flare combustion in todays final rule.
With regard to the definition of flare combustion efficiency, EPA agrees that it will be less
confusing to reporters to clarify the use of the term natural gas in the definition. In todays
final rule, the definition of flare combustion efficiency is the fraction of hydrocarbon gas, on a
volume or mole basis, that is combusted at the flare burner tip.

Organization:
((98.230 (2)) Onshore petroleum and natural gas production includes the verbiage all EOR
operations using CO2. We propose that EOR operations using CO2 be defined with facility
definitions and that facility definitions are consistent with the reporting requirements of other
facilities. The current statement could be interpreted such that all wellsites, test sites and
1702
processing facilities would be grouped into one large facility under the definition of a production
facility. This would cause unnecessary reporting burden on companies practicing Enhanced Oil
Recovery that would be inconsistent with the reporting burden of natural gas facilities
Response: EPA disagrees that EOR operations should be identified with an individual facility
definition. EPA views EOR simply as an operating practice in crude oil production involving
certain unique processes and stream compositions. Todays final rule has been clarified in
98.232(c) that in onshore production GHG emissions from sources of EOR injection pump
blowdown and EOR hydrocarbons liquids dissolved CO2 are included in subpart W.
Commenter:
Comment on 98.6: Definitions
Condensate
Current air quality permitting programs have more specific definitions of condensate to
prevent a production water tank from being defined as a condensate storage tank.
Revision: "Condensate means hydrocarbon liquids that remain liquid at standard conditions (68
degrees Fahrenheit and 29.92 inches Mercury) and are formed by condensation from, or
produced with, natural gas, and which have an American Petroleum Institute gravity ("API
gravity") of 40 degrees or greater."
Response: EPA disagrees that it should eliminate water from the definition of condensate. For
more information, see the response to EPA-HQ-OAR-2009-0923-1040-12.

Commenter:
Flare Combustion
CO2 and N2O are not "unburned hydrocarbons" or products of incomplete combustion
Revision: "Flare combustion means the burning of hydrocarbons and flammable gases in flares."
Response: EPA agrees that the definition of flare combustion emissions was confusing. In
todays final rule the term has been modified to flare stack emissions and clarifies that it means
CO2 and N2O from partial combustion of hydrocarbon gas sent to a flare plus CH4 emissions
1703
resulting from the incomplete combustion of hydrocarbon gas in flares. Please see the response
to EPA-HQ-OAR-2009-0923-0847-4.

Commenter:
Gas to oil ratio (GOR)
Revision for clarification.
Revision: "The gas/oil ratio (GOR) is the ratio of the volume of gas that comes out
of solution, to the volume of oil at standard conditions."
Response: EPA revised the final rule methods using GOR in 98.233(l) and 98.233(m). Please

Optical Gas Imaging Instrument
Section 98.234 requires the use of an "optical gas imaging instrument" for annual leak surveys.
This is an undefined term.
Response: EPA disagrees that the term optical gas imaging instrument is insufficiently
described for reporters to be able to comply with subpart W. Please see the response to EPA
HQ-OAR-2009-0923-0049-2.

In addition, the requirements appear to target continuous bleed pneumatic controllers, as defined
in Section 98.7. The definition for High-Bleed Pneumatic Devices should be modified as
follows: High-Bleed Pneumatic Devices Controllers are automated continuous bleed flow
control devices powered by pressurized natural gas and used for maintaining a process condition
such as liquid level, pressure, delta-pressure and temperature. Part of the gas power stream which
is regulated by the process condition flows to a valve actuator or other process controller. Highbleed controllers where it vents (bleeds) to the atmosphere at a rate in excess of six standard
cubic feet per hour.
1704
Response: Todays final rule clarifies the definitions of pneumatic devices as continuous highbleed, continuous low-bleed, and intermittent.

The definition for Low-Bleed Pneumatic Devices should be modified as follows:
Low-Bleed Pneumatic Devices Controllers mean pilot operated automated flow control devices
powered by pressurized natural gas and used for maintaining a process condition such as liquid
level, pressure, delta-pressure and temperature. Part of On actuation, the gas power stream which
is regulated by the process condition flows to a valve actuator or other process controller. On deactuation, the gas power stream used to actuate the process controller where it vents (bleeds) to
the atmosphere. Although the amount of gas vented at pilot operated controllers depends on the
amount of power gas used for actuation, the typical rate is at a rate equal to or less than six
standard cubic feet per hour.

Acid Gas means hydrogen sulfide (H2S) and carbon dioxide (CO2) contaminants that are
separated from sour natural gas by an acid gas removal.
The word unit is missing from the end of the definition. API requests replacing the EPA
definition with the following: The hydrogen sulfide and/or carbon dioxide contained in or
extracted from gas or other streams (from Gas Processors Suppliers Association).
Acid Gas Removal Vent Stack Emissions mean the acid gas separated from the acid gas
absorbing medium (e.g., an amine solution) and released with methane and other light
hydrocarbons to the atmosphere or a flare.
The definition only mentions CH4 emissions, omitting CO2 emissions from this source. This is
inconsistent with the requirements to report CO2 emissions and not CH4 emissions under Section
98.233(d).
1705
Air injected flare means a flare in which air is blown into the base of a flare stack to induce
complete combustion of low Btu natural gas (i.e., high non-combustible component content).
API requests the following changes to this definition: A flare in which air is blown into the base
of a flare stack to induce complete combustion. Remove, of low Btu natural gas (i.e. high
noncombustible component content.) It is unnecessary to include the Btu content of the gas
stream being burned in the flare. At times high Btu gas could be burned in an air injected flare to
provide enough oxygen to promote complete combustion.
Blowdown vent stack emissions mean natural gas released due to maintenance and/or blowdown
operations including but not limited to compressor blowdown and emergency shut-down (ESD)
system testing.
This definition of a blowdown vent stack emissions seems to exclude compressors in CO2
service. This exclusion is most likely an oversight. Also this definition might wrongly include
emissions that are not traditionally considered blowdown emissions by broadly applying the
definition to all emissions released due to maintenance activities. Blowdown emissions are
specific and the definition should be more precise as in the example below.
Blowdown vent stack emissions means natural gas or carbon dioxide trapped within a
compressor and associated piping that is released during start-up and shut down procedures for
that compressor.
Compressor means any machine for raising the pressure of a natural gas by drawing in low
pressure natural gas and discharging significantly higher pressure natural gas.
Modify the definition as follows: Compressor means any machine for raising the pressure of
natural gas or CO2 by drawing in low pressure natural gas or CO2 and discharging significantly
higher pressure natural gas or CO2
Condensate means hydrocarbon and other liquid separated from natural gas that condenses due
to changes in the temperature, pressure, or both, and remains liquid at storage conditions,
includes both water and hydrocarbon liquids.
API requests the following alternative definition: Liquid formed by the condensation of a liquid
or gas; specifically, the hydrocarbon liquid separated from natural gas because of changes in
temperature and pressure when the gas from the reservoir was delivered to the surface separators.
Such condensate remains liquid at atmospheric temperature and pressure.
Dehydrator vent stack emissions means natural gas released from a natural gas dehydrator
system absorbent (typically glycol) reboiler or regenerator, including stripping natural gas and
motive natural gas used in absorbent circulation pumps.
API requests the definition be revised to clarify emissions from the dehydrator to the atmosphere
or a flare are subject to the rule but not emissions routed back to the process. API recommends
the definition be revised to state natural gas released from a natural gas dehydrator system
absorbent (typically glycol) reboiler or regenerated to the atmosphere or a flare, including
1706
Desiccant means a material used in solid-bed dehydrators to remove water from raw natural gas
by adsorption. Desiccants include activated alumina, palletized calcium chloride, lithium
chloride and granular silica gel material. Wet natural gas is passed through a bed of the granular
or pelletized solid adsorbent in these dehydrators. As the wet gas contacts the surface of the
particles of desiccant material, water is adsorbed on the surface of these desiccant particles.
Passing through the entire desiccant bed, almost all of the water is adsorbed onto the desiccant
material, leaving the dry gas to exit the contactor.
This definition combines two entirely different types of gas drying. Separate definitions should
be provided for the two types of desiccant material used for dehydration in the oil and gas sector.
One is the mole-sieve type desiccant system that uses alumina, silica, etc., where the wet gas is
passed through and the water is adsorbed onto the surface of the desiccant materials. Once
loaded, the beds are regenerated by passing hot gas through them to drive the water out.
Generally, a plant/facility will have at least two sets of mole sieve type vessels so one can be in
service and one in regeneration or cooling in preparation for switching beds. Unless the facility is
using hot nitrogen for regeneration, the medium is generally hot gas and is recycled back to the
inlet of the plant. This type lasts for many years before the beads have to be replaced.
The second is a deliquescent (sacrificial) desiccant type that uses pelletized salts like calcium
chloride or lithium chloride. They work by the hygroscopic salt attracting the water and being
dissolved to a saturated brine. These are the type that must be opened for
replacement/replenishment of the desiccant.
The two new definitions that should be provided for deliquescent and mole sieve desiccant are as
follows:
Desiccant - deliquescent means a hydro-phillic salt material used in solid-bed dehydrators to
remove water from raw natural gas by adsorption and dissolving the salt into a brine.
Deliquescent desiccants include, but are not limited to, pelletized calcium chloride or lithium
chloride. Wet natural gas is passed through a bed of the granular or pelletized material where
water is adsorbed onto the desiccant material and dissolves it into a brine, leaving the dry gas to
exit the contactor. When the majority of the salt material is dissolved, the dehydrator vessel must
be opened and the salt replenished.
Desiccant - mole sieve means a material used in solid-bed dehydrators to remove water from raw
natural gas by adsorption. Desiccants include, but are not limited to, activated alumina and
granular silica gel material. Wet natural gas is passed through a bed of the granular or pelletized
solid adsorbent in these dehydrators. As the wet gas contacts the surface of the particles of
desiccant material, water is adsorbed on the surface of these desiccant particles. Almost all the
water passing through the entire desiccant bed is adsorbed onto the desiccant material, leaving
the dry gas to exit the contactor. When the adsorption media is water loaded it is regenerated by
passing hot gas through the media to dry it and prepare it for subsequent use. The regeneration
gas is typically captured and recycled into the process for hydrocarbon gas or vented if nitrogen
is used for regeneration. Mole sieve desiccants have a long life and typically are not replaced on
1707
a frequent schedule.
Dehydrator means a device in which a liquid absorbent (including but not limited to desiccant,
ethylene glycol, diethylene glycol, or triethylene glycol) directly contacts a natural gas stream to
absorb water vapor.
Modify the definition as follows: Dehydrator means a device in which a liquid absorbent
(including but not limited to desiccant, ethylene glycol, diethylene glycol, or triethylene glycol)
or desiccant directly contacts a natural gas stream to absorb water vapor.
Enhanced Oil Recovery (EOR) means the use of certain methods such as water flooding or gas
injection into existing wells to increase the recovery of crude oil from a reservoir. In the context
of this rule, EOR applies to injection of critical phase carbon dioxide into a crude oil reservoir to
enhance the recovery of oil.
Modify the definition as follows: Enhanced Oil Recovery (EOR) means the use of certain
methods such as water flooding or gas injection into existing wells to increase the recovery of
crude oil from a reservoir. In the context of this rule, EOR applies to injection of critical phase
carbon dioxide into a crude oil reservoir to enhance the recovery of oil.
Field means standardized field names and codes of all oil and gas fields identified in the United
States as defined by the Energy Information Administration Oil and Gas Field Code Master List.
The preamble notes that EPA proposes to incorporate by reference the American Association of
Petroleum Geologists (AAPG) Committee on Statistics of Drilling (CSD) Geologic Code
Provinces Code Map, not the EIA Field Code Master List.
Flare combustion means unburned hydrocarbons including CH4, CO2, N2O emissions resulting
from the incomplete combustion of gas in flares.
The definition of flare combustion does not appear to make sense. It seems the term should be
flare combustion emissions not just flare combustion. Also unburned hydrocarbons are only
part of the resulting emissions, as NOx, CO, and SOx emissions are also emitted. Modify the
definition as follows: Flare combustion emissions means unburned hydrocarbon emissions,
including CH4, resulting from the incomplete combustion of gas in flares along with CO2 and
N2O formed in the combustion process.
Flare combustion efficiency means the fraction of natural gas, on a volume or mole basis, that is
combusted at the flare burner tip.
Suggest an alternative definition: means the fraction of carbon in the flared stream, on a volume
or molar basis, that oxidizes to CO2 through combustion.
Fugitive emissions means those emissions which are unintentional and could not reasonably pass
through a stack, chimney, vent or other functionally-equivalent opening.
1708
This differs from existing federal rules such as 40 CFR 52.21(i)(20) and 40 CFR 63.2, which
state Fugitive emissions means those emissions that could not reasonably pass through a stack,
chimney, vent, or other functionally equivalent opening. The fugitive emissions definition
should not result in an expansion of the current CAA regulatory definition of fugitive emissions
by including the phrase "unintentional" emissions.
Gas gathering/booster stations mean centralized stations where produced natural gas from
individual wells is co-mingled, compressed for transport to processing plants, transmission and
distribution systems, and other gathering/booster stations which co-mingle gas from multiple
production gathering/booster stations. Such stations may include gas dehydration, gravity
separation of liquids (both hydrocarbon and water), pipeline pig launchers and receivers, and gas
powered pneumatic devices.
API requests that the definition state Compressors on an individual well site, with single or
multiple wells, and serving only the wells on the same site are excluded from this definition.
New Definition: Gas well: A well completed for production of natural gas from one or more gas
zones or reservoirs. Such wells contain no completions for the production of crude oil (from the
Energy Information Administration).
High-Bleed Pneumatic Devices are automated flow control devices powered by pressurized
natural gas and used for maintaining a process condition such as liquid level, pressure, deltapressure and temperature. Part of the gas power stream which is regulated by the process
condition flows to a valve actuator controller where it vents (bleeds) to the atmosphere at a rate
in excess of six standard cubic feet per hour.
This definition implies the "high-bleed" pneumatic device and the controller are different things,
when in actuality the "high-bleed" pneumatic device/controller and the valve actuator are
different devices. In addition, the correct term for these devices is pneumatic controllers. The
definition should also be clarified that it is intended to include only continuous bleed devices, as
described in Section 98.233(a). The following revisions are suggested:
High-Bleed Pneumatic Devices Controllers are automated "continuous bleed" flow control
devices powered by pressurized natural gas and used for maintaining a process condition such as
liquid level, pressure, delta-pressure and temperature. Part of the gas power stream which is
regulated by the process condition flows to a valve actuator or other process controller. Highbleed controllers where it vents (bleeds) to the atmosphere at a rate in excess of six standard
Low-Bleed Pneumatic Devices mean automated flow control devices powered by pressurized
equal to or less than six standard cubic feet per hour. The definition should be clarified that it is
intended to include only continuous bleed devices, as defined in Section 98.233(b). The
following revisions are suggested:
1709
is regulated by the process condition, flows to a valve actuator or other process controller. On
de-actuation, the gas power stream used to actuate the process controller where it vents (bleeds)
to the atmosphere. Although the amount of gas vented at pilot operated controllers depends on
the amount of power gas used for actuation the typical rate is at a rate equal to or less than six
New Definition: Natural gas distribution means distribution pipelines (not interstate pipelines or
intrastate pipelines) and metering and regulating stations, that physically deliver natural gas to
end users. This category excludes infrastructure and pipelines delivering natural gas directly to
major industrial users upstream of the local distribution company inlet and "farm taps" upstream
of the local distribution company inlet.
New Definition: Oil well means a well completed for the production of crude oil from at least
one oil zone or reservoir (from the Energy Information Administration).
Onshore petroleum and natural gas production owner or operator means the entity who is the
permitee to operate petroleum and natural gas wells on the state drilling permit or a state
operating permit where no drilling permit is issued by the state, which operates an onshore
petroleum and/or natural gas production facility (as described in Section 98.230(b)(2). Where
more than one entity are permitees on the state drilling permit, or operating permit where no
drilling permit is issued by the state, the permitted entities for the joint facility must designate
one entity to report all emissions from the joint facility.
Modify the definition as follows: Onshore petroleum and natural gas production owner or
operator means the entity who is the permitee to operate petroleum and natural gas wells on the
state or federal drilling permit or a state or federal operating permit where no drilling permit is
issued by the state or federal agency having jurisdiction, which operates an onshore petroleum
and/or natural gas production facility (as described in Section 98.230(b)(2). Where more than
one entity are permitees on the state drilling permit, or operating permit where no drilling permit
is issued by the state, the permitted entities for the joint facility must designate one entity to
report all emissions from the joint facility.
Re-condenser means heat exchangers that cool compressed boil-off gas to a temperature that will
condense natural gas to a liquid.
Modify the definition as follows: Re-condenser means heat exchangers that cool compressed
boil-off gas from LNG to a temperature that will condense natural gas to a liquid.
Sales oil means produced crude oil or condensate measured at the production lease automatic
custody transfer (LACT) meter or custody transfer meter tank gauge.
Modify the definition as follows: Sales oil means produced crude oil or condensate measured at
1710
the production lease automatic custody transfer (LACT) meter or by custody transfer meter tank
gauging.
Sour natural gas: means natural gas that contains significant concentrations of hydrogen sulfide
and/or carbon dioxide that exceed the concentrations specified for commercially saleable natural
gas delivered from transmission and distribution pipelines.
API requests that EPA clarify that this definition only applies to this rule.
Turbine meter means a flow meter in which a gas or liquid flow rate through the calibrated tube
spins a turbine from which the spin rate is detected and calibrated to measure the fluid flow rate.
Modify the definition as follows: Turbine meter means a flow meter in which a gas or liquid
flow rate through the calibrated tube meter spins a turbine from which the spin rate is detected
and calibrated to measure converted to the fluid flow rate. Turbine meters are typically factory
calibrated and can only be field calibrated by use of a "proving meter" approach.
Well workover means the performance of one or more of a variety of remedial operations on
producing oil and gas wells to try to increase production. This process also includes high-rate
back-flow of injected water and sand used to re-fracture and prop-open new fractures in existing
low permeability gas reservoirs.
Modify the definition as follows: Well workover means the performance of operations on
producing oil and gas wells to increase production.
Response: EPA agrees and has revised todays final rule definition of acid gas by inserting
unit and referring to hydrogen sulfide and/or carbon dioxide. .
EPA disagrees that the definition of acid gas removal vent emissions does not already include
CO2. The definition states that it is vented acid gas in addition to methane and other light
hydrocarbons. Acid gas is defined as hydrogen sulfide and/or carbon dioxide in the rule.
EPA agrees with the suggested clarification to the terms air-injected flare and compressor.
Todays final rule reflects these clarifications.
EPA agrees that it is clearer for reporters to revise the definition of blowdown vent stack
emissions to include CO2. Additionally, EPA disagrees that specifying compressor blowdown
activities for start up and shutdown activities will encompass the desired blowdown activities.
EPA disagrees with the proposed changes to the definition of condensate; please see the response
to EPA-HQ-OAR-2009-0923-1018-19.
EPA disagrees that it is not clear in the current definition of dehydrator vent that gas which is
captured and recycled back into the system is not included in reporting under subpart W. Please
1711
EPA did not intend to include mole sieve dehydration. For further details, please see the
With regard to the definition of enhanced oil recovery, the commenter suggested the exact same
definition, verbatim, that is currently in the rule.
EPA disagrees that the commenters proposed alternative definition to flare combustion clarifies
the term better for the purposes of subpart W. Please see the response to EPA-HQ-OAR-2009
0923-0847-4. However, EPA does agree that the definition provided in the proposed rule could
be improved, and todays final rule defines flare combustion efficiency as the fraction of
hydrocarbon gas, on a volume or mole basis, that is combusted at the flare burner tip.
EPA agrees that it should alter the definition of fugitive emissions to not include the term
unintentional to conform to the definition in other rules. EPA has partially addressed this issue
by replacing the term fugitive emission with equipment leak in todays final rule, excluding
the term unintentional from the definition.
With regard to the definition of gathering and boosting stations, EPA has decided not to include
gathering and boosting equipment at this time. For more information, please see Section II.F of
EPA agrees, and has revised todays final rule in 98.230 to clarify that farm taps are not included
in the definition of natural gas distribution under subpart W. Please see the response to EPA
HQ-OAR-2009-0923-1016-24.
EPA agrees, and has revised the rule to include definitions for gas well and oil well in 98.238.
EPA disagrees with the comments on onshore petroleum and natural gas production owner or
operator. However EPA has revised this definition in todays final rule in 98.238.
With regard to the proposed definition changes to high-bleed and low-bleed pneumatic devices,
EPA has altered the way in which pneumatic devices are defined. Please see the response to
EPA-HQ-OAR-2009-0923-1024-38.
EPA does not consider the suggested changes to the definition of re-condenser or turbine meter
to be substantially different or clarified over what is already in subpart W; and thus, disagrees
that the changes are necessary.
EPA agrees, and has revised todays final rule to strike the term meter from the phrase
custody transfer meter tank gauge in the definition of sales oil.
EPA disagrees that it must clarify that the definition of sour natural gas applies only to this rule
as the definition of sour gas is in 98.6 and therefore applies to the entire MRR.
EPA disagrees that the simplified definition of well workover proposed by the commenter
provides a useful clarification for reporters. EPA has determined that its more detailed
description provides a clearer understanding of the activities that are included as a well
workover.
1712

Section 98.231(b) Reporting threshold. The term wellhead should be clarified to specifically
refer to production wellhead. Natural gas storage fields have wellheads and there is no
indication throughout the subpart that EPA intended this for storage wellheads.
Response: EPA disagrees that changing the term wellhead to production wellhead will
provide any practical clarification to subpart W. Additionally, since storage wellheads are
covered in underground storage, EPA is referring to both types of wellheads. However, todays
final rule clarifies that stationary or portable combustion emissions be reported under subpart W
from onshore production and natural gas distribution.

Definitions
The definition for fugitive emissions at 98.6 should exclude the term unintentional because
not all fugitive emissions are unintentionally emitted. For example, known leaking valves that
are not repaired are intentional emission sources.
The definitions for conventional wells and unconventional wells at 98.6 should be revised
to be consistent with industry terminology. For example, unconventional wells may include tight
sands, coalbed methane and gas shale production, oil shales, tar sands, etc. The use of fracture
treatment is not the only determinant of whether a well is conventional or unconventional. It is
our expectation that all types of petroleum development, both conventional and unconventional,
will be covered under Subpart W, and the definitions should reflect that expectation.
The definition of the offshore petroleum and natural gas production source category at
98.230 should be revised to eliminate the clause affixed temporarily or permanently to the
offshore submerged lands. All GHG emissions generated from offshore petroleum and natural
gas production activities should be accounted for and reported. This should not just be limited to
the period of time a rig is anchored in place.
Response: EPA has revised the definition of fugitive emissions to equipment leaks. Please
With regard to the definitions for conventional and unconventional wells, EPA agrees with this
comment has revised the definitions Please see the response to EPA-HQ-OAR-2009-0923-1040
13.
1713
EPA does not agree to extend reporting under subpart W to mobile drilling rigs in offshore
production. See response to comment EPA-HQ-OAR-2009-0923-1201-17 for additional
information.

EPA Should Clearly Define the Term "Leaker" in the Final Rule.
Dow comments that EPA should clearly define the term "Leaker" in the final rule. EPA should
define the term if one uses an optical gas imaging instrument and should also define the term if
one uses an OVA/TVA as suggested in the comment above. Dow comments that the definition of
leaker should be a concentration above 10,000 ppmv if measured by an OVA/TVA. A level of
10,000 ppmv is consistent with the level used in the Clearstone study to determine "leakers" and
"non-leakers".
Response: EPA agrees that it should state what constitutes a leak using each of the detection
alternatives it now allows. Todays final rule clarifies that any leak that can be seen through an
optical gas imaging instrument is considered a leak, and for Method 21 leak detection
instruments, it conforms to the definition of leak provided in Method 21. For more information
on the alternatives methods of leak detection, please see Section II.F of the preamble to todays
final rule.

The Dehydrator Vent Stack Emissions definition states that it includes the motive natural gas
used by the circulation pump. The natural gas employed and expelled by a circulation pump does
not vent through the dehydrator vent stack; therefore, it is not accurate to group this stream with
the emissions from the dehydrator still vent. In addition, natural gas driven pneumatic pump
venting is a distinct source type that Subpart W addresses separately from dehydrator vent
stacks. We request that EPA remove circulation pump motive natural gas from the dehydrator
vent stack emissions definition.
The Fugitive Emissions Detection definition states that it includes emissions from equipment,
components, and other point sources. Fugitive emissions is a term that naturally excludes point
sources; therefore, other point sources should not be included in the definition of fugitive
emissions. For example, if reciprocating compressor rod packing or centrifugal compressor wet
seal degassing emissions vent through a piped vent, they are not a fugitive emission source;
1714
however, if they do not pass through a piped vent, they are a fugitive emission source. We
request that EPA remove other point sources from the fugitive emissions detection
definition.
The Sales Oil definition states that it can be measured at a lease automatic custody transfer
(LACT) meter or custody transfer meter tank gauge. A tank gauge, not a meter, measures sales
oil at the tank. We request that EPA remove the term meter from the term custody transfer
meter tank gauge in the sales oil definition.
Response: EPA agrees that some types of dehydrator circulation pumps do not vent the motive
gas through the reboiler vent stack, and so todays final rule removes the term stack from the
dehydrator vent emissions definition. Todays final rule requires reporting of glycol dehydrator
vent emissions, which will include the reboiler vent stack and in that vent, glycol circulation
pump pneumatic gas when an energy-assist pump which commingles rich TEG containing gas
under contactor pressure with additional wet gas for mechanical advantage of the pump driver
(piston). For dehydrators that use other types of gas pneumatic circulation pumps that vent the
pneumatic gas directly to the atmosphere, those emissions will be reported as part of the glycol
dehydrator emissions. Todays final rule requires the use of a software program, such as the
Glycalc program, for all dehydrators of 0.4 million cubic feet per day and above, which
includes emissions from a gas pneumatic pump. EPA disagrees that these types of glycol
circulation pumps would be covered under the natural gas driven pneumatic pump source
category. Pneumatic driven pumps refer to chemical injection pumps and other such production
well-site pumps that inject chemicals such as methanol into wells, flowlines, etc.
With regard to the definition of fugitive emissions detection todays final rule replaces the term
fugitive with equipment leaks and defines this the same way as the proposed rule defined
fugitives, including other point sources. This is because the rule requires use of the IR camera
for identifying such point sources as vents on transmissions compressor station condensate tanks,
indicating the possibility of through-leaking scrubber dump valves.
In todays final rule EPA has clarified the definition to state that sales oil means produced crude
oil or condensate measured at the production lease automatic custody transfer meter or custody
transfer tank gauge.

This section requires operators to report emissions separately for standby equipment.
However, the term standby is not defined anywhere in the rule (neither Subpart W nor Subpart
A). Moreover, equipment is generally not specifically designated by operators as primary or
standby. This adds an additional layer of reporting that is unwarranted and confusing, uses
nebulous terminology, and provides no meaningful information. We request that EPA remove
this requirement.
1715
Response: EPA has determined that most reporters understand the term standby with regard to
compressors. Todays final rule uses the term standby in context with the mode of operation
in which a compressor is not operating, but could be started-up at any moment. The functional
terms used in todays final rule, generically referred to as standby are: not operating,
pressurized and not operating, depressurized. Examples include compressors at a natural gas
transmission compressor station which are started up for peak demand periods, or feed and
residue gas compressors in processing facilities that allow rotation of compressors for
maintenance without losing facility throughput.
With regard to the requirement that reporters must measure emissions from compressors in all
modes of operation, todays final rule clarifies that compressors are only to be measured in the
operating mode in which they are found so long as each compressor is measured in the notoperating, depressurized mode at least once every three years. For more information, please see
the response to Section II.F of the preamble to todays final rule and response to comment EPA
HQ-OAR-2009-0923-0055-16.

Consistent with the comment on 98.233(a), we request that EPA replace the word count with
estimate.
98.236(c)(2):
Consistent with the comment on 98.236(c)(18)(i), we request that EPA replace the word
count with estimate.
98.236(c)(3):
Consistent with the comment on 98.233(c), we request that EPA replace the word count with
estimate.
Response: EPA disagrees that it should replace the data reporting requirement to report the
count of pneumatic devices and pneumatic pump counts with estimates. In order to estimate
greenhouse gas emissions from these sources, reporters must use the counts of each piece of
equipment in the equations listed under the methodologies, EPA requires this information as
necessary activity data to inform future policy.

1716

Also, the SUGF recognizes the usage of the term "large compressors," in Table W-4 of the
preamble, and other locations within the proposed rule. As with other CAA programs, large
compressors should be defined by EPA, to include compressors of a certain horsepower (hp)
level. Thus, the SUGF recommends that EPA further define "large compressors" to be classified
as compressors with a level of 2,000 hp or more.
Response: EPA no longer uses the term large compressor to classify compressors under
subpart W, please see EPA-HQ-OAR-2009-0923-1019-2.
The Condensate definition states that it includes both water and hydrocarbon liquids. Although a
condensate storage tank in a production field typically includes both water and condensate, the
two separate into different phases and are removed from the tank as separate streams. The tank
operator documents the condensate production volume as the condensate phase that is removed
from the tank. It is not accurate to state that condensate includes water, and IPAMS requests that
EPA remove water from the condensate definition.
Response: EPA disagrees that it should remove water from the definition of condensate for the
purposes of Subpart W. Please see the response to EPA-HQ-OAR-2009-0923-1040-12.

The Conventional wells definition states that they do not employ hydraulic fracturing.
Conventional gas reservoirs are typically underground formations composed of sandstone,
limestone, dolomite, and/or mixed mineralogy, and conventional wells are those that produce
from these types of reservoirs. The term conventional does not necessarily consider whether
the formation has been hydraulically fractured. Many of the conventional and
unconventional wells in the United States have been hydraulically fractured. IPAMS requests
that EPA define the terms conventional wells and unconventional wells based on the wells
reservoir type, as opposed to whether the well operates in a fractured reservoir.
Response: EPA does not agree to define conventional and unconventional by reservoir type.

1717

The Dehydrator vent stack emissions definition states that it includes the motive natural gas used
by the circulation pump. The natural gas employed and expelled by a circulation pump does not
vent through the dehydrator vent stack; therefore, it is not accurate to group this stream with the
emissions from the dehydrator still vent. In addition, natural gas driven pneumatic pump venting
is a distinct source type that Subpart W addresses separately from dehydrator vent stacks.
IPAMS requests that EPA remove circulation pump motive natural gas from the dehydrator
vent stack emissions definition.
Response: EPA disagrees that it should remove gas-assisted glycol circulation pumps from the
dehydrator vent stack emissions source, please see the response to EPA-HQ-OAR-2009-0923
1206-29.

Fugitive emissions: EPA should define fugitive emissions in proposed Subpart W consistent with
existing rules promulgated under the CAA to avoid confusion among various regulatory
programs. EPA should apply the definition of fugitive emissions already found in 40 CFR
52.21(i)(20) and 40 CFR 63.2, which state that fugitive emissions means those emissions that
could not reasonably pass through a stack, chimney, vent or other functionally equivalent
opening. For example, leaks from natural gas driven pneumatic controller and valves, pump
seals, non-pneumatic pumps and connectors should all be considered fugitive emissions. The use
of an additional term, such as equipment leak, will only introduce additional confusion and
increase the reporting burden without providing additional useful information to EPA. IPAMS
also requests that EPA remove other point sources from the fugitive emissions detection
definition.
Response: EPA agrees to modify the proposed definition for fugitive emissions, and has deleted
the word unintentional in todays final rule. EPA has replaced the term fugitive emissions
with equipment leaks. This change was made to ensure consistency with the terminology in the
Alternative Work Practice to Detect Leaks from Equipment for 40 CFR parts 60, 63, and 65.
While the commenter suggested that the addition of a term like equipment leaks would cause
confusion, EPA disagrees that it will cause confusion in place of (rather than in addition to) the
term fugitive emissions.
With regard to the suggested alteration to the term fugitive emissions detection, please see the
1718

The Gas to oil ratio (GOR) definition states that it is the ratio of the volume of gas at standard
temperature and pressure produced from a volume of oil when depressurized to standard
temperature and pressure. The oil and gas industry does not typically measure GOR in this
manner. GOR is usually expressed as the measured natural gas volume that separates at actual
separator temperature and pressure (corrected to standard temperature and pressure) divided by
the measured oil volume from the storage tank(s). IPAMS requests that EPA modify the GOR
definition to accurately reflect the current industry-accepted method.
0923-1040-14.

The Reservoir definition states that it is characterized by a single natural gas pressure system.
This statement is not accurate because a reservoir can have multiple pressure systems that are
separate pods with permeability barriers between them. IPAMS requests that EPA remove the
last sentence from this definition.
Response: EPA agrees with the comment. For more information, see response to comment EPA
HQ-OAR-2009-0923-1040-15.
The Sales oil definition states that it can be measured at a lease automatic custody transfer
(LACT) meter or custody transfer meter tank gauge. A tank gauge, not a meter, measures sales
oil at the tank. IPAMS requests that EPA remove the term meter from the term custody
transfer meter tank gauge in the sales oil definition.
Response: EPA agrees, and has revised todays final rule to remove the term meter from the
phrase custody transfer meter tank gauge in the definition of sales oil.
1719

The Sour natural gas definition states that it is natural gas that contains significant concentrations
of H2S and/or CO2 that exceed the concentrations specified for commercially saleable natural
gas. The concentration of CO2 does not determine whether natural gas is considered sour. The
oil and gas industry considers natural gas to be sour based only on the H2S concentration.
Although CO2 content is indirectly controlled by heat content specifications, most natural gas
contracts do not contain a specification for CO2 content unless the natural gas contains an
excessive concentration of CO2. In addition, the concentration of H2S at which a natural gas
stream is considered sour is not a standard value across the industry. This concentration varies
according to the production area and the sales contract. IPAMS requests that the sour natural
gas definition not reference CO2 content, and suggests that EPA define the concentration at
which natural gas is consistently considered sour to be 100 ppmv or more.
Response: EPA disagrees that it should remove CO2 from the definition of the term sour gas.
This term is used only in the definitions to describe the purpose of an acid gas removal unit,
which applies equally to produced natural gas containing quantities of H2S and/or CO2 which
must be reduced to meet sales gas quality. The methods for estimating emissions from acid gas
removal units do not depend on the definition of sour natural gas; only on the concentration of
CO2 in the feed, product or vent streams.
With regard to defining one consistent level at which natural gas is considered to be sour, EPA
disagrees that this is necessary. EPA is not setting standards on the gas for the purposes of this
rule. EPA is only concerned with the sour natural gas that is treated in an acid gas removal
unit because the operator must either remove CO2, H2S, or both in order to get the gas to pipeline
specifications. EPA is concerned because this is a significant source of greenhouse gas
emissions. The threshold at which natural gas is considered to be sour does not affect
compliance with this subpart.

The Sweet Gas definition states that it is natural gas with low concentrations of H2S and/or CO2.
As discussed in the comment for Sour natural gas, the concentration of CO2 does not determine
whether natural gas is considered sour or sweet. IPAMS requests that the sweet gas
definition not reference CO2 content.
Response: EPA disagrees with this comment. For further details, please see the response to
EPA-HQ-OAR-2009-0923-1298-18.

1720

The Unconventional wells definition states that they are wells that employ hydraulic fracturing.
The term unconventional applies to wells that produce from reservoirs that are not
conventional, most often coal beds and shales. In addition, the term unconventional typically
refers to wells that produce from tight formations with a permeability less than 0.1 milli-Darcies.
The term unconventional does not necessarily consider whether a reservoir has been
hydraulically fractured. Many of the conventional and unconventional wells in the United
States have been fractured. IPAMS requests that EPA define the terms conventional wells and
unconventional wells based on the wells reservoir type, as opposed to whether the well
operates in a fractured reservoir.
Response: EPA agrees, and has revised todays final rule for the definitions of completions and
workovers and the associated reporting requirements. Please see the response to EPA-HQ-OAR
2009-0923-1040-13.

The Well completions definition states that it is a process that allows for the flow of petroleum or
natural gas from newly drilled wells to expel drilling and reservoir fluids and test the reservoir
flow characteristics. This definition is not accurate. The term well completion typically refers
to the process of connecting the well bore to the reservoir. It may include treating the formation
or installing tubing, packer(s), or lifting equipment. IPAMS requests that EPA modify the well
completions definition to reflect the description given here.
Response: EPA agrees, and has revised todays final rule for the definition of well completion
such that it reflects the activities discussed in this comment. In todays final rule, the definition
adds additional clarification that completions are the process of connecting the well bore to the
reservoir and may include treating the formation or installing tubing, packer(s), or lifting
equipment.

The Well workover definition states This process also includes high-rate back-flow of injected
water and sand used to re-fracture and prop open new fractures in existing low permeability gas
reservoirs. IPAMS requests that EPA modify this sentence to read, This process may also
1721
include high-rate flowback of injected gas, water, oil, and sand used to fracture or re-fracture and
prop open new fractures in existing lower permeability gas reservoirs.
Response: EPA revised the final rule for the definition of well workover in such a way that
partially matches the commenters proposal. For more information see response to comment
EPA-HQ-OAR-2009-0923-1040-18.

98.6 Definitions
BP suggests the following changes to the definitions (proposed language changes/additions are
shown in green):
- Acid Gas means hydrogen sulfide (H2S) and carbon dioxide (CO2) contaminants that are
separated from sour natural gas by an acid gas removal unit.
- Air injected flare means a flare in which air is blown into the base of a flare stack to induce
complete combustion. of low Btu natural gas (i.e., high non-combustible component content).
Compressor means any machine for raising the pressure of a natural gas or CO2 by drawing in
low pressure natural gas or CO2 and discharging significantly higher pressure natural gas or
CO2.
- Condensate means hydrocarbon and other liquid separated from natural gas that condenses due
to changes in the temperature, pressure, or both, and remains liquid at storage conditions ,
includes both water and hydrocarbon liquids.
Desiccant means a material used in solid-bed dehydrators to remove water from raw natural gas
by adsorption. Desiccants include activated alumina, palletized calcium chloride, lithium
chloride and granular silica gel material. Wet natural gas is passed through a bed of the granular
or pelletized solid adsorbent in these dehydrators. As the wet gas contacts the surface of the
material, leaving the dry gas to exit the contactor.
This definition combines two entirely different types of gas drying. Separate definitions should
be provided for the two types of desiccant material used for dehydration in the oil and gas sector.
One is the mole-sieve type desiccant system that uses alumina, silica, etc. where the wet gas is
passed through and the water is adsorbed onto the surface of the desiccant materials. Once
loaded, the beds are regenerated by passing hot gas through them to drive the water out.
1722
Generally, a plant/facility will have at least two sets of mole sieve type vessels so one can be in
service and one in regeneration or cooling in preparation for switching beds. Unless the facility is
using hot nitrogen for regeneration the medium is generally hot gas and is recycled back to the
inlet of the plant. This type lasts for many years before the vessel needs to be opened and beads
have to be replaced.
The second is a deliquescent (sacrificial) desiccant type that uses pelletized salts like calcium
chloride or lithium chloride. They work by the hygroscopic salt attracting the water and being
dissolved to saturated brine. These are the type that must be opened for
replacement/replenishment of the desiccant.
The two new definitions that should be provided for deliquescent and mole sieve desiccant are as
follows:
Desiccant - deliquescent means a hydro-phillic salt material used in solid-bed dehydrators to
remove water from raw natural gas by adsorption and dissolving the salt into a brine.
Deliquescent desiccants include, but are not limited to, pelletized calcium chloride or lithium
chloride. Wet natural gas is passed through a bed of the granular or pelletized material where
water is adsorbed onto the desiccant material and dissolves it into a brine, leaving the dry gas to
exit the contactor. When the majority of the salt material is dissolved, the dehydrator vessel must
be opened and the salt replenished.
Desiccant - mole sieve means a material used in solid-bed dehydrators to remove water from
raw natural gas by adsorption. Desiccants include, but are not limited to, activated alumina and
granular silica gel material. Wet natural gas is passed through a bed of the granular or
pelletized solid adsorbent in these dehydrators. As the wet gas contacts the surface of the
material, leaving the dry gas to exit the contactor. When the adsorption media is water loaded it
is regenerated by passing hot gas through the media to dry it and prepare it for subsequent use.
The regeneration gas is typically captured and recycled into the process for hydrocarbon gas or
vented if nitrogen is used for regeneration. Mole sieve desiccants have a long life and typically
are not replaced on a frequent schedule.
- Flare combustion means unburned hydrocarbons, including CH4, emissions resulting from
incomplete combustion of gas in flares along with CO2, and N2O formed in the combustion
process emissions resulting from the incomplete combustion of gas in flares.
The definition of flare combustion does not appear to be correct. It seems the term should be
flare combustion emissions not just flare combustion. Also unburned hydrocarbons are only
part of the resulting emissions, as NOx, CO, and SOx emissions are also emitted.
- Fugitive emissions means those emissions which are unintentional and could not reasonably
pass through a stack, chimney, vent or other functionally-equivalent opening.
This differs from existing federal rules such as 40 CFR 52.21(i)(20) and 40 CFR 63.2, which
1723
state Fugitive emissions means those emissions that could not reasonably pass through a stack,
chimney, vent, or other functionally equivalent opening. The fugitive emissions definition
should not result in an expansion of the current CAA regulatory definition of fugitive emissions
by including the phrase "unintentional" emissions.
- Gas gathering/booster stations mean centralized stations where produced natural gas from
individual wells is co-mingled, compressed for transport to processing plants, transmission and
distribution systems, and other gathering/booster stations which co-mingle gas from multiple
production gathering/booster stations. Such stations may include gas dehydration, gravity
separation of liquids (both hydrocarbon and water), pipeline pig launchers and receivers, and gas
powered pneumatic devices.
The definition should be modified to include: Compressors on an individual well site or at a
centralized tank battery or flow station, with single or multiple wells, and serving only the wells
on or connected to the same site are excluded from this definition.
- New Definition for Gas Well: EPA uses the term gas well throughout Subpart W but does not
offer a definition for this term and thus creates confusion. For example It is unclear whether the
calculations required for gas wells in 98.233(f), (g), and (h) would apply to wells that do not
produce natural gas in commercially saleable quantities. BP recommends a definition consistent
with the intent of the rule.
New Definition: Gas well: A well completed for production of natural gas from one or more gas
zones or reservoirs. Such wells contain no completions for the production of crude oil.
High-Bleed Pneumatic Devices are automated flow control devices powered by pressurized
in excess of six standard cubic feet per hour.
This definition implies the "high-bleed" pneumatic device and the controller are different things,
when in actuality the "high-bleed" pneumatic device/controller and the valve actuator are
different devices. In addition, the correct term for these devices is pneumatic controllers. The
definition should also be clarified that it is intended to include only continuous bleed devices, as
described in 98.233(a). The following revisions are suggested:
High-Bleed Pneumatic Devices Controllers are automated "continuous bleed" flow control
devices powered by pressurized natural gas and used for maintaining a process condition such as
liquid level, pressure, delta-pressure and temperature. Part of the gas power stream which is
regulated by the process condition flows to a valve actuator or other process controller. Highbleed controllers where it vents (bleeds) to the atmosphere at a rate in excess of six standard
- Low-Bleed Pneumatic Devices mean automated flow control devices powered by pressurized
natural gas and used for maintaining a process condition such as liquid level, pressure, delta
1724
pressure and temperature. Part of the gas power stream which is regulated by the process
equal to or less than six standard cubic feet per hour.
The following revisions are suggested:
is regulated by the process condition, flows to a valve actuator or other process controller. On
de-actuation, the gas power stream used to actuate the process controller where it vents (bleeds)
to the atmosphere. Although the amount of gas vented at pilot operated controllers depends on
the amount of power gas used for actuation the typical rate is at a rate equal to or less than six
- New Definition: BP recommends a definition of natural gas distribution be added to the rule as
follows:
Natural gas distribution means distribution pipelines (not interstate pipelines or intrastate
pipelines) and metering and regulating stations, that physically deliver natural gas to end users.
This category excludes infrastructure and pipelines delivering natural gas directly to major
industrial users upstream of the local distribution company inlet and "farm taps" upstream of the
local distribution company inlet.
- New Definition for Oil Well: BP recommends a definition of oil well be added to the rule as
follows:
Oil well means a well completed for the production of crude oil from at least one oil zone or
reservoir.
- Onshore petroleum and natural gas production owner or operator means the entity who is the
permitee to operate petroleum and natural gas wells on the state or federal drilling permit or a
state or federal operating permit where no drilling permit is issued by the state or federal agency
having jurisdiction, which operates an onshore petroleum and/or natural gas production facility
(as described in Section 98.230(b)(2). Where more than one entity are permitees on the state
drilling permit, or operating permit where no drilling permit is issued by the state, the permitted
entities for the joint facility must designate one entity to report all emissions from the joint
facility.
- Re-condenser means heat exchangers that cool compressed boil-off gas from LNG to a
temperature that will condense natural gas to a liquid.
- Sales oil means produced crude oil or condensate measured at the production lease automatic
custody transfer (LACT) meter or by custody transfer meter tank gauging.
- Turbine meter means a flow meter in which a gas or liquid flow rate through the calibrated tube
1725
meter spins a turbine from which the spin rate is detected and calibrated to measure converted to
the fluid flow rate. Turbine meters are typically factory calibrated and can only be field
calibrated by use of a "proving meter" approach.
B. Section 98.7(m)(l)
Add the following AAPG publication to the reference in 98.7(m)(1): Alaska Geologic Province
Boundary Map; AAPG - CSD - USGS; 1978
C. Section 98.230(a)(1) Offshore petroleum and natural gas production
The rule text includes inland waters which is not correct. In addition, 98.230(a)(1) and Section
98.238 both provide the definition of offshore petroleum and natural gas production. The
definitions differ in that Section 98.238 states [a]ll production equipment that is connected via
causeways or walkways are one facility. This text should be added to Section 98.230(a)(1). BP
recommends Section 98.230(a)(1) be revised to:
Offshore petroleum and natural gas production is any platform structure, affixed temporarily or
permanently to offshore submerged lands, that houses equipment to extract hydrocarbons from
beneath the ocean or lake floor and that transfers such hydrocarbons to hydrocarbon to storage,
transport vessels, or onshore. In addition, offshore production includes secondary platform
structures and storage tanks connected to the platform structure via causeways or walkways.
D. Section 98.230(a)(4) Onshore natural gas transmission compression
The definition is not clear regarding residue gas compression located at gas plants. Modify the
rule text as follows:
Onshore natural gas transmission compression. Onshore natural gas transmission compression
means any fixed combination of compressors that move natural gas at elevated pressure from
production fields or natural gas processing facilities, in transmission pipelines, to natural gas
distribution pipelines, or into storage. In addition, a transmission compressor station may
includes equipment for liquids separation, natural gas dehydration, and tanks for the storage of
water and hydrocarbon liquids. Residue (sales) gas compression located at Natural Gas
Processing Facilities are included in the processing category and excluded from this category.
E. Section 98.230(a)(8) Natural Gas Distribution
The description does not exclude direct delivery to major industrial users and/or farm taps
from classification as distribution. Modify the description as follows: Natural gas distribution
means distribution pipelines (not interstate pipelines or intrastate pipelines) and metering and
regulating stations, that physically deliver natural gas to end users. This category excludes
infrastructure and pipelines delivering natural gas directly to major industrial users upstream of
the local distribution company inlet and "farm taps" upstream of the local distribution company
inlet.
1726
Response: EPA agrees, and has revised todays final rule for the definition of acid gas; please
EPA agrees with the suggested clarification to the terms air-injected flare and compressor.
Todays final rule reflects these clarifications.
EPA disagrees with the proposed changes to the definition of condensate; please see the response
to EPA-HQ-OAR-2009-0923-1018-19.
With regard to the term desiccant dehydrator, EPA intended that only desiccant dehydrators
using a hydro-phillic salt material are included under subpart W, and thus, mole sieve
dehydration is not included.
EPA disagrees the proposed definition for flare combustion; however, however EPA has revised
todays final rule for this definition. Please see the response to EPA-HQ-OAR-2009-0923-0847
4.
EPA agrees that it should alter the definition of fugitive emissions to not include the term
unintentional to conform to the definition in other rules. Please see the response to EPA-HQ
OAR-2009-0923-1151-17.
EPA agrees, and has revised todays final rule to strike the term meter from the phrase
custody transfer meter tank gauge in the definition of sales oil.
In todays final rule, EPA has decided not to include gathering and boosting equipment at this
time. For more information, please see Section II.F of the preamble to todays final rule.
EPA agrees, and has revised the rule to include definitions for gas well and oil well in 98.238.
With regard to the proposed definition changes to high-bleed and low-bleed pneumatic devices,
EPA has altered the way in which pneumatic devices are defined. Please see the response to
EPA-HQ-OAR-2009-0923-1024-38.
EPA does not consider the suggested changes to the definition of re-condenser or turbine meter
to be substantially different or clarified over what is already in subpart W; and thus, disagrees
with that the changes are necessary.
EPA agrees with the addition of the AAPG publication to Section 98.7 and has made the
necessary changes to todays final rule.
EPA disagrees with the comments on onshore petroleum and natural gas production owner or
operator. However, EPA has revised this definition in todays final rule in Section 98.238.
EPA disagrees with the commenter on the exclusion of inland waters from offshore and has
retained it in the source category definition in Section 98.230 in todays final rule. EPA agrees
with the commenter regarding walkways, and has revised todays final rule offshore source
category definition in Section 98.230. EPA has also revised the definition of offshore in Section
98.238, and clarifies that lakes are included.
1727
EPA agrees with the comment about onshore natural gas transmission compression. For further
EPA agrees and has revised todays final rule in Section 98.230, to clarify that farm taps are not
included in the definition of natural gas distribution in todays final rule For further details,

EPA Seriously Underestimates the Burdens on Gas Utilities
It is difficult to assess the economic burden of collecting and reporting this information, given
that EPA has not provided a clear definition of several key terms, including optical gas
scanning equipment and M&R stations.
Response: EPA disagrees that it has not provided a clear definition of optical gas imaging
equipment. Todays final rule prescribes that reporters may use an optical gas imaging
instrument in accordance with 40 CFR part 60, subpart A, section 60.18(i)(1) and (2). These are
the same instruments used for the Alternative Work Practice for Method 21.
EPA agrees, and has revised todays final rule to clarify which M&R stations are required for
leak detection. For further details, please see the comment response in Section II.F of the

Based on Canadian LDC reporting experience, the proposed facility definition for LDCs appears
to be workable, although we have had considerable discussion relating to how to split
"transmission" from "distribution" for companies that operate and are regulated as a single entity.
Response: EPA has reviewed these comments and todays final rule provides guidance on colocated facilities. Normally, transmission systems are regulated by a different entity than
distribution facilities, and the separation of those facilities is clear from the regulations.
However, it may be possible for a local distribution company public utility agreement to cover
equipment normally found in transmission compressor stations. This would be considered a colocated facility. For example, if a local distribution company is operating within its facility
boundaries and under a single regulatory agreement that covers the distribution operations as
1728
well as one or more transmission compressor stations, todays final rule requires the reporter to
determine the industry segment for which the majority of emissions occur and report all
equipment within that facility for which there is a method defined. A reporter must still report
all emissions sources from each co-located source. For example, if distribution and transmission
facilities are co-located, all emissions sources for both distribution and transmission must be
reported. If the data collection or measurement methodologies are not consistent between source
categories, the reporter uses the methodologies as defined under the source category from which
the emissions are greatest. For more information, see response to comment EPA-HQ-OAR-2009
0923-1024-14.

The Final Rule should clarify that customer meter sets are part of the service line and, hence, are
not to be reported in any other category.
The terms .meter. .regulator, .meters, and .regulator stations. (or .M and R stations.) are not
defined, and frequently used interchangeably, in the Proposed Rule. M and R Station is a terms
of art used in the gas distribution industry to refer to gate stations where utilities receive gas from
their pipeline-suppliers. Regulator stations include district regulators that cut the pressure of gas
at various points throughout the distribution piping network. The preamble makes it clear that
EPAs intent is to follow the industry usage:
In the distribution segment, high-pressure gas from natural gas transmission pipelines enters a
.city gate, station, which reduces the pressure and distributes the gas through primarily
underground mains and service lines to individual end users. Distribution system CH4 and CO2
emissions result mainly from fugitive emissions from above ground gate stations (metering and
regulating stations), below grade vaults (regulator stations), and fugitive emissions from buried
pipelines. At gate stations, fugitive and vented CH4 emissions primarily come from valves, openended lines, connectors, pressure safety valves, and natural gas driven pneumatic devices.
Individual customer meter sets are, by industry norm and Federal pipeline safety regulations,
considered part of a customer service line and, are obviously not part of a gate station or a
district regulator. However, customer meter sets usually contain both a regulator and a meter.
Because the terms .meter, M and R Station, and .regulator, are not defined in the Proposed Rule,
these terms could be unreasonably interpreted to include over 60 million customer meter sets.
Such an interpretation would impose an enormous undue burden on LDCs. Almost all customer
meter sets are above ground and, by operation of Section 98.232(i)(1) and 98.233(q) of the
Proposed Rule, would require an annual leakage survey to determine the count of such meters
that are leaking. Accordingly, the treatment of customer meter sets as above ground meters,
regulators or M and R stations would impose enormous costs and undue burdens.
1729
Such massive additional costs and burdens would be entirely unjustified. As noted, industry
norm is to treat customer meter sets as part of a service line. Notably, PHMSA regulations
confirm this practice at they include customer meter sets within the definition of a service line.
Thus, any fugitive emissions from customer meter sets are already included in the Population
Emission Factor for service lines in Table W-7. Accordingly, to include customer meter sets
again as separate meters, regulators, or M&R stations, would be to provide a double count of
emissionsat an enormous unjustified expense. For these reasons, APGA urges EPA to clarify
that the terms, .meter, .regulator, and .M and R station, as used in the Proposed Rule do not
include customer meter sets.
Response: EPA agrees, and has revised todays final rule to clarify which M&R stations are
required for leak detection. For further details, please see the comment response in Section II.F
of the preamble to todays final rule.
EPA disagrees that residential meters are accounted for in the service line population emission
factors provided in subpart W. These emission factors were developed in the EPA/GRI 1996
report which provides separate emission factors for the residential meters they represent only
the service pipeline leaks. Subpart W does not require the reporting of residential meters, only
the service lines.
M&R Definition
The proposed rules failure to define M&R stations makes it very difficult to determine
compliance requirements and ensure that the necessary resources to support compliance are
scheduled and/or obtained. The rule does not clearly state if end-use residential, commercial and
industrial customer meter sets are to be included in the annual survey. For Xcel Energy, the
number of stations requiring leak surveys could increase from 2800, if just district regulating
stations are included, to over 1.8 million if the definition of M&R stations includes all sales
meters for industrial, commercial, and residential customers. The resources and time required for
performing the leak surveys and reporting differ substantially depending upon the M&R
definition employed. Since natural gas in distribution systems is required to be odorized and our
customers readily call if they detect any odors, Xcel Energy recommends that the M&R
definition only include district regulating stations.
required for leak detection, and customer meters are not included. For further details, please see
the comment response in Section II.F of the preamble to todays final rule.

1730

Clarify the Subpart A Definitions of Connector and Open-ended valves or lines As Applied
to Natural Gas Facilities Under Subpart W
The 2010 Proposal would require LDCs to report fugitive emissions from and conduct annual
leak surveys of eight listed components at city gates and M&R stations, including connectors
and open ended lines. 260 Proposed section 98.233(q) requires an annual leak detection survey of
these components. 75 Fed. Reg. at 18637, 18643.] In Subpart A of the MRR, EPA defined the
terms connector and open ended lines as follows:
Connector means the flanged, screwed, or other joined fittings used to connect pipe line
segments, tubing, pipe components (such as elbows, reducers, Ts or valves) or a pipe line
and a piece of equipment or an instrument to a pipe, tube or piece of equipment. A common
connector is a flange. Joined fittings welded completely around the circumference of the
interface are not considered connectors for the purpose of this part.
Open-ended valve or lines (OELs) means any valve, except pressure relief valves, having one
side of the valve seat in contact with process fluid and one side open to atmosphere, either
directly or through open piping. 261
Under these definitions, a standard LDC regulator and meter could be deemed to have up to 90
connectors and several open-ended lines. In addition, it could include virtually any regulator or
meter. Almost all components (meters, regulators, valves) in a distribution system are connected
into the system by bolted flanges on both sides of the component. One problem we have with the
generic definition in Subpart A is that it does not give our members guidance regarding flanges.
EPA could receive reports regarding two categories of leaks associated with each component
(i.e. component flanges and the component itself). It is not clear to us whether this is what the
agency is seeking.
The reference to process fluid in the term open-ended valve or line is confusing. In theory,
under fluid mechanics, a fluid is defined as either a liquid or gas. There are some circumstances
in which natural gas can behave as a liquid (for example when converted to LNG under
extremely low temperatures). Nevertheless, in the context of natural gas distribution or
underground storage facilities, natural gas is in a gaseous state and is not a liquid. In the normal
sense of the word liquid, natural gas distribution systems do not transport fluids other than
occasional, very small quantities of hydrocarbon condensates or water. Accordingly, under a
common sense interpretation of the Subpart A definition, LDCs do not have any open ended
valves or lines. Given that EPA specifically listed this component for LDCs and other natural gas
260
Proposed 40 C.F.R. 98.232(i)(1) requires LDC to report fugitive emissions from connectors, block valves,
control valves, pressure relief valves, orifice meters, other meters, regulators, and open ended lines at above
ground meter regulators and gate station[s].
261
40 C.F.R. 98.6.
1731
segments, we suspect the agency really means process fluid or gas and the term is intended to
apply to permanent purge vents. We request that EPA provide an amended definition of openended line for purposes of Subpart W that includes this clarifying change..
To limit the burden of leak surveys and reporting for city gates and district M&R stations and to
facilitate compliance and implementation of the rule, AGA asks EPA to adopt a different
definition of these terms for use in Subpart W that will better align with natural gas distribution
operations. We request that EPA adopt the following definitions for use in Subpart W:
Connector means flanged, screwed, or other joined fittings used to connect pipe line segments,
tubing, such as elbows, reducers, Ts or pipe unions. A common connector is a flange. Joined
fittings welded completely around the circumference of the interface are not considered
connectors for the purpose of this part. This definition does not include flanges required to install
components such as valves, meters, regulators, and open-ended lines.
Open-ended valve or lines (OELs) means any valve, except pressure relief valves, having one
side of the valve seat in contact with process fluid or gas and one side open to atmosphere, either
directly or through open piping.
Response: EPA disagrees that it must clarify for subpart W that the term fluid in the subpart
A definition for open-ended line to expressly state that gas is also considered a fluid. It is clear
in the context of natural gas distribution that the process fluid is natural gas and uncapped or
unplugged vent and drain connections are the open-ended lines subject to leak surveys in the
rule. Further, todays final rule defines a size of over 0.5 inches diameter subject to leak surveys,
so it would be extremely unlikely that one LDC meter and regulator would have up to 90
connectors and several open-ended lines. Most tubing and tubing connectors will be under the
size limits.
EPA disagrees that it should change the definition of connector to not include flanges required to
install components such as valves, meters, regulators, and open-ended lines. Todays final rule
requires custody transfer city gate stations to perform a leak survey, and the leaks found to be
categorized for application of leaker factors provided in the rule to estimate emissions. Flanges
that connect valves and meters to piping are connectors. Seal welded flanges found leaking are
still leaking connectors. Valve bonnet flanges, stem packing, grease nipples and other
components found leaking on a valve exclusive of the piping connector flanges are valve leaks.
The tables of leak factors by component type are taken from recent equipment leak surveys
where the contractor had to combine leaks in logical categories as is required in todays final
rule.

In the proposed Subpart W Rule, EPA requires fugitive emissions reported from "open ended
1732
pipe". We believe that the EPA means "open ended pipe" to be "vent stacks for relief valves or
blow down stacks" at meter and regulator stations. The MDU LDCs do not normally inventory
"vent stacks or blow down stacks" and MDU recommends that the EPA either clarify what this
term means or remove it from the Subpart W Rule. Also, the EPA requires fugitive emissions
reported from "connectors" at meter and regulator stations. MDU does not know exactly what
this term encompasses. Also, the MDU LDCs do not inventory all connectors, such as 90 degree
elbows, T-fittings, small sections of pipe and small control valves.
Response: Todays final rule does not limit equipment leaks from open-ended lines to
blowdown vent stacks (subpart W does not use the term open ended pipe). The definition for
open-ended lines is provided in subpart A, Open-ended valve or lines (OELs) means any valve,
except pressure relief valves, having one side of the valve seat in contact with process fluid or
gas and one side open to atmosphere, either directly or through open piping. Hence, an
uncapped or unplugged vent or drain connection is an open-ended line. An equipment blowdown
valve connected to an open vent stack is an open-ended line during the time that the blowdown
valve is closed. Special provisions are described in the rule for calculating or measuring
equipment blowdown emissions when blowdown valves are opened, during which time they are
not open-ended lines.
Todays final rule requires LDCs to perform leak surveys at above ground custody transfer city
gate stations, and use leaker factors for all leaks found in component categories. For more
information, see response to comment EPA-HQ-OAR-2009-0923-1016-22.

Clarify the Definition of Distribution Facility To Exclude Transmission Lines and Small,
Remote Distribution Systems Operated by LDCs
AGA recommends that EPA revise the definition of Natural Gas Distribution Facility to clearly
exclude natural gas transmission pipelines, referring to the definitions of distribution and
transmission provided in DOT PHMSAs regulations at 49 C.F.R. Part 192, 192.3. In the
2010 Proposal for Subpart W, EPA proposes to define a distribution facility as distribution
pipelines, metering stations, and regulating stations that are operated by a Local Distribution
Company (LDC) that is regulated as a separate operating company by a public utility
commission or that are operated as an independent municipally-owned distribution system.
While we believe your intent is to exclude intrastate and interstate transmission pipelines, the
definition is inadvertently unclear, because some LDCs operate significant amounts of
transmission lines. A clear definition of distribution line that excludes gathering lines and
transmission lines as defined in at 49 C.F.R.192.3 will remove that ambiguity.
Most LDCs are served by affiliate or interstate transmission pipelines, and the definition of a
distribution facility as an LDC, which is regulated as a separate company by a Public Utilities
1733
Commission (PUC) is designed to fit that business model. However, several AGA members
operate both transmission and distribution systems. The definition of a distribution facility needs
to be clarified to determine the specific operations that are included in it.
The definition of a distribution facility should be based on physical attributes such as Maximum
Allowable Operating Pressure (MAOP) instead of corporate ownership or regulatory body in
order to make the distinction between distribution facilities and other facilities clearer.
In addition, small, remote distribution sub-systems should be exempted from the reporting
requirements, even if owned/operated by the same large company, because the amount of
emissions identified from these sub-systems would not be commensurate with the time, cost, and
effort required to report them. Our members estimate that small sub-systems serving fewer than
140,000 customers emit far less than the 25,000 metric tons of CO2e threshold. We therefore
propose that distribution facility should be defined as a contiguous network of distribution
lines and services operating at less than 100 psig Maximum Allowable Operating Pressure
(MAOP) and serving more than 140,000 customers. In addition, the term distribution line
should be defined to have the same meaning as in 49 C.F.R. 192.3.
Response: EPA agrees that it would be beneficial to reporters to clarify the definition of natural
gas distribution to not include transmission pipelines. Todays final rule defines natural gas
distribution in 98.230 as the distribution pipelines (not interstate pipelines or intrastate pipelines)
and metering and regulating stations that physically deliver natural gas to end users. This
category does not include customer meters and the infrastructure and pipelines delivering natural
gas directly to major industrial users and "farm taps" upstream of the local distribution company
inlet. EPA disagrees with revising the facility definition of natural gas distribution to include a
threshold for operating pipeline pressure and number of customers. EPA determined that the
estimated number of natural gas distribution facilities reporting under subpart W and the
estimated burden are reasonable. Please see the response to comment EPA-HQ-OAR-2009
0923-0049-7 and the Economic Impact Analysis Section 5 in EPA-HQ-OAR-2009-0923.

Clarify Distribution Facility Includes Only The Portion of Gate Stations Owned by LDCs,
Excludes Customer Meters, and Excludes Farm Taps
In the proposed rule, gate stations are considered to be part of the distribution system, and LDCs
are required to report emissions from them. However, in many cases, LDCs only own part of the
gate stations, which are also considered to be part of the transmission system. AGA therefore
urges EPA to:
- Clarify that only the distribution portion of the gate station is within the natural gas distribution
system for purposes of the threshold determination and subject to reporting by the LDC;
1734
- Clarify that metering and/or regulation to customers is not subject to the leak survey and
reporting requirements;
- If the intention is to include larger customer meter and regulator stations, then a clearly defined
threshold should be identified, such as meter design capacity (e.g., 15,000 Dth/day) or nominal
regulator size (e.g., four inch); and
- Clarify that farm taps (i.e. gas pressure regulation installed on services connected to
transmission pipelines) are not included in the count of metering and/or regulator stations.
Response: EPA agrees that it is useful to clarify for which distribution meter and regulator
stations it prescribes leak detection and leaker factors. Todays final rule clarifies that customer
meters are not covered by subpart W, and leak detection is required only at above ground
metering and regulator city gate stations at which custody transfer occurs. The remainder of
above ground metering and regulator stations will be monitored by a population count and a
population emission factor developed from the average emissions per station that was monitored
with leak detection. For more information, please see the comment response in Section II.F of
EPA did not intend for reporters to monitor and report emissions from farm taps. In todays final
rule in Section 98.230, the definition of natural gas distribution clarifies that farm taps are not
included.
EPA disagrees that it must clarify that only the distribution portion of the gate station is required
for monitoring by natural gas distribution facilities. Distribution facilities are only to monitor the
prescribed sources under their ownership or operation. That is, distribution facilities do not
report the transmission portion of the gate station that is not part of the distribution system.

Section 98.230(a)(8) Natural Gas Distribution. The description does not exclude direct delivery
to major industrial users and/or farm taps from classification as distribution. Modify the
description as follows: Natural gas distribution means distribution pipelines (not interstate
pipelines or intrastate pipelines) and metering and regulating stations, that physically deliver
natural gas to end users. This category excludes infrastructure and pipelines delivering natural
gas directly to major industrial users upstream of the local distribution company inlet and "farm
taps" upstream of the local distribution company inlet.
Response: EPA agrees, and has revised todays final rule to clarify that farm taps are not
included in the definition in Section 98.230 of natural gas distribution under subpart W. Please
1735

Facility Definitions and Threshold Determination Issues
A. The Definition of Facility for Distribution is Overbroad
EPA is proposing to cast aside the normal, common-sense definition of facility in the case of
natural gas distribution. The definition used for decades in other Clean Air Act programs is the
same as the definition provided in Subpart A of the GHG Mandatory Reporting Rule (MRR).
The term facility is defined in 40 C.F.R. 98.6 as follows:
Facility means any physical property, plant, building, structure, source, or stationary equipment
located on one or more contiguous or adjacent properties in actual physical contact or separated
solely by a public roadway or other public right-of-way and under common ownership or
common control, that emits or may emit any greenhouse gas.
We support EPAs proposal to use this traditional definition of facility for natural gas pipeline
compression, underground storage, LNG storage, and LNG import/export facilities that our
members operate.
However, we are troubled by the novel, expansive definition EPA proposes for natural gas
distribution. Instead of using the traditional definition of facility for natural gas distribution, EPA
is now proposing to define a gas distribution facility to encompass virtually the entire
distribution system operated by a single company, which in some cases can span an entire state.
Proposed section 98.238 defines the term as follows:
75 Fed. Reg. at 18647.
As proposed, it is also not clear whether the definition would encompass the distribution pipes,
city gates and M&R stations located in multiple states if they are operated by the same company.
We do not believe this was the agencys intent. If the intent is to cut the boundary of the
distribution facility at the state border, then at a minimum, the definition should be revised to
say that the term natural gas facility means the listed pipe and equipment located within a single
state that are operated by an LDC and regulated by that states public utility commission.
However, the normal definition of the term facility should be used throughout the MRR
limited to sites with contiguous boundaries. This type of facility-level reporting by owners and
operators would be consistent with other Clean Air Act and state-level regulatory programs,
1736
which will facilitate compliance and minimize the administrative burden of the proposed rule.
B. In the Alternative, EPA Should Clarify that the Unusual Definition of Facility for GHG
Reporting Purposes Should Not Be Used for the PSD Tailoring Rule or Other Clean Air Act
Purposes
AGA is particularly concerned that the expansive definition of distribution facility could be
adopted in other contexts, such as future phases of the Prevention of Significant Deterioration
(PSD) Tailoring Rule for greenhouse gas emissions. If EPA does not adopt the traditional
definition of facility for natural gas distribution, then the agency should at least clarify that
other Clean Air Act programs should continue to use the traditional definition of facility to
determine whether emissions from a facility exceed regulatory thresholds requiring pre
construction permits or the application of control technology. Otherwise, far too many operating
decisions within a local distribution system could require time and resource consuming
regulatory decisions.
Response: EPA disagrees that it must redefine the natural gas distribution facility definition in
Section 98.238 of todays final rule to adhere more closely to the definition of facility in
subpart A of The Final Mandatory GHG Reporting Rule (Final MRR), (40 CFR part 98). For
the purposes of todays final rule, the provided definitions supersede those provided in subpart
A. EPA deemed it necessary to define facilities this way in order to achieve the desired 80%
coverage at the 25,000 tonne CO2e threshold. For more information regarding EPAs facility
definition choice, please see chapter 4.c.i. and 5. of the background technical support document
in the rule making docket (EPA-HQ-OAR-2009-0923).
In todays final rule, the definition of facility, as it relates to natural gas distribution, has been
clarified to be the distribution pipelines, metering stations, and regulating stations that are
operated by a Local Distribution Company (LDC) that is regulated as a separate operating
company by a public utility commission or that are operated as an independent municipalityowned distribution system. EPA disagrees that the definition should include state boundaries, if
an LDC serves areas on both sides of a state boundary so long as it is regulated as a single
operating company then it will be considered one facility by this rule. EPA understands the
concern that two systems separated by several states under common ownership being considered
a single facility; however, it has determined that the current definition appropriately addresses
that concern.
EPA disagrees that it should stipulate in subpart W any conditions or exclusions on the data
collected, which is expressly for the purpose of informing future policy without bias on what or
how that policy may be determined. The data cuts both ways by not only indicating sources of
emissions large enough to warrant abatement, but also sources of emissions too small to costeffectively abate, or that current practices are adequately managing those emissions.

1737

Revise Facility Definition for Underground Storage to Clearly Exclude Transmission Lines
Between Wellheads
AGA appreciates that EPA apparently has not defined underground storage facilities quite as
broadly as it has defined onshore natural gas production facilities which would extend across
multiple states in an entire hydrocarbon basin as defined by the American Association of
Petroleum Geologists which is assigned a three digit Geologic Province Code pursuant to
proposed section 98.238. However, it appears the definition of underground storage facility
could be nearly as broad, could cross state lines and would extend across many square miles. The
source category for underground natural gas storage is defined to include widely scattered
surface facilities connected to an underground storage salt cavern or depleted gas or oil reservoir.
This broad definition raises a concern similar to one that we raised with respect to distribution.
Underground storage facilities include transmission lines. The DOT reclassified all gathering
lines in natural gas storage facilities as transmission lines several years ago. Only lines
connecting production wells to processing and compression stations may be called gathering
lines. In the 2010 Proposal, EPA proposes to exclude transmission lines from Subpart W,
however, it appears that the agency only contemplated this exclusion in the context of interstate
natural gas pipeline systems and did not realize that DOT-defined transmission lines are also
operated by LDCs and within underground storage facilities. AGA asks that EPA revise section
98.230(a)(5), defining the source category for underground natural gas storage, to explicitly
exclude transmission lines.
Response: EPA disagrees that it must clarify the definition of underground natural gas storage
to not include transmission lines, because transmission pipelines are not required for monitoring
and reporting under subpart W as specified under Sections 98.232 - GHGs to report or 98.233 Calculating GHG emissions of todays final rule.

Clarify the definition of natural gas distribution facility to clearly delineate between transmission
and distribution assets;
Clarify the definition of above grade meter regulators (M&R) stations to exclude customer
meters to reduce reporting costs;
Response: EPA agrees, and has revised todays final rule to clarify the delineation between
transmission and distribution assets in 98.230. Please see the response to EPA-HQ-OAR-2009
0923-1016-23.
EPA agrees that it is beneficial to reporters to clarify that above ground meter and regulator
stations do not include customer meters. Todays final rule states that customer meters are not
1738
included for reporting under subpart W. Please see the comment response in Section II. F of the

Distribution Facility Definition
The proposed definition of facility for natural gas distribution differs from the definition of
facility applied in the remainder of the MRR. For natural gas distribution facility, EPA is
proposing the following definition:
The Clean Energy Group generally agrees with this definition for the purposes of this rule
especially since this is the approach LDCs currently utilize for other reporting programs
including the Department of Transportations (DOT) Pipeline and Hazardous Material Safety
Administration, the Federal Energy Regulatory Commission (FERC), and EPAs Natural Gas
Star Program.
However, many LDCs also own and operate transmission assets in addition to their distribution
assets. As such, this facility definition should be further developed in order to delineate the
boundary between the distribution and transmission assets. The Clean Energy Group also
suggests that distribution facilities be redefined to exclude small, remote sub-systems in order to
reduce reporting costs as noted below.
Response: EPA revised todays final rule to clarify the delineation between transmission and
distribution assets in 98.230. Please see the response to EPA-HQ-OAR-2009-0923-1016-23. It
is unclear to EPA what the commenter meant by remote sub-systems.

Metering & Regulating (M&R) Stations
The definition of "Metering & Regulating (M&R) Stations" is used differently in different
sections of the regulation and is not clearly defined. In the preamble (pg 18617) you use the term
"gas metering and pressure regulation equipment (M&R)". In 98.232.i.l, you refer to "meter
regulators" and in the other sections, including 98.236, for example, you refer to Metering and
1739
Regulating (M&R) stations. Not only is the term used differently in the proposed rule but it is
unclear whether M&R includes only city gate meters, many of which do not have meters (and
only have regulators), or whether it also includes customer meters (all of our residential meters
have both a regulator and a meter)? In the preamble (pg 18617) it states that customer meters are
part of the distribution system and that an LDC must "report for all the distribution facilities that
they own or operate". Given this description, our interpretation is M&R would include customer
(commercial and residential) meters.
Response: Todays final rule clarifies the definition of M&R stations. Please see the comment
response in Section II. F of the preamble to todays final rule.

Definition of source category: Natural gas distribution
The definition of LDC in 98.230 includes metering and regulating stations that physically deliver
NG to end users. There is another definition of LDC in section 98.238. The definitions are not
the same. EPA should refer to one or the other or define them the same in order to be consistent.
Does the natural gas distribution source category include customer meters?
Response: EPA disagrees that it provides separate definitions for LDC in the proposed rule.
The definition in 98.230 is the source category definition for natural gas distribution while the
definition in 98.238 is the facility definition for natural gas distribution.
Todays final rule clarifies that reporters are not required to monitor and report emissions data
for customer meters. Please see the comment response in Section II.F of the preamble to todays
final rule.

Section 98.230(a) (4) Onshore natural gas transmission compression. The definition is not clear
regarding residue gas compression located at gas plants. Modify the rule text as follows:
Onshore natural gas transmission compression means any fixed combination of compressors
that move natural gas at elevated pressure from production fields or natural gas processing
facilities, in transmission pipelines, to natural gas distribution pipelines, or into storage. In
addition, a transmission compressor station may includes equipment for liquids separation,
natural gas dehydration, and tanks for the storage of water and hydrocarbon liquids. Residue
(sales) gas compression located at Natural Gas Processing Facilities are included in the
processing category and excluded from this category.
1740
Response: EPA agrees in part with the proposed edits to the onshore natural gas transmission
compression definition stated in this comment. Todays final rule corrects the grammar error in
the definition with the additional clarification that residue gas compressors owned and operated
by a gas processing facility, and dedicated to moving sales gas from that gas processing facility
to a transmission system, whether inside or outside the processing facility fence line, are a part of
the gas processing facility.

Revise Facility Definition for Underground Storage
Consumers appreciates that EPA apparently has not defined underground storage facilities quite
as broadly as it has defined onshore natural gas production facilities which would extend
across multiple states in an entire hydrocarbon basin as defined by the American Association of
Petroleum Geologists which is assigned a three digit Geologic Province Code pursuant to
proposed section 98.238. However, it appears the definition of underground storage facility
could be nearly as broad, could cross state lines and would extend across many square miles. The
source category for underground natural gas storage is defined to include widely scattered
surface facilities connected to an underground storage salt cavern or depleted gas or oil reservoir.
For the reasons stated above, AGA urges EPA to revise section 98.230(a)(5) to use a more
traditional definition of facility for underground natural gas storage.
Response: EPA disagrees that it should revise the definition of underground storage in todays
final rule. The definition for underground natural gas storage includes the processes and
operations (including compression, dehydration and flow measurement), and all the wellheads
connected to the compression units located at the facility that inject and reproduce natural gas.

Clarify the Definition of Distribution Facility To Exclude Transmission Lines and Small,
Remote Distribution Systems Operated by LDCs
Consumers recommends that EPA revise the definition of Natural Gas Distribution Facility to
clearly exclude natural gas transmission pipelines, referring to the definitions of distribution
and transmission provided in DOT PHMSAs regulations at 49 C.F.R. Part 192, 192.3. In the
2010 Proposal for Subpart W, EPA proposes to define a distribution facility as distribution
pipelines, metering stations, and regulating stations that are operated by a Local Distribution
1741
Company (LDC) that is regulated as a separate operating company by a public utility

commission or that are operated as an independent municipally-owned distribution system.
While we believe your intent is to exclude intrastate and interstate transmission pipelines, the
definition is inadvertently unclear, because some LDCs operate significant amounts of
transmission lines. A clear definition of distribution line that excludes gathering lines and
transmission lines as defined in at 49 CFR192.3 will remove that ambiguity. In addition, the
term distribution line should be defined to have the same meaning as in 49 C.F.R. 192.3.
Clarify Distribution Facility Includes Only Portion of Gate Stations Partly Owned by LDCs,
Excludes Customer Meters and Excludes Farm Taps
In the proposed rule, gate stations are considered to be part of the distribution system, and LDCs
are required to report emissions from them. However, in many cases, LDCs only own part of the
gate stations, which are also considered to be part of the transmission system. Therefore,
Consumers urges EPA to:
Clarify that only the distribution portion of the gate station is within the natural gas distribution
system for purposes of the threshold determination and subject to reporting by the LDC;
Clarify that metering and/or regulation to customers is not subject to the leak survey and
reporting requirements;
If the intention is to include larger customer meter and regulator stations, then a clearly defined
threshold should be identified, such a meter design capacity (e.g., 15,000 Dth/day) or nominal
regulator size (e.g., four inch); and
Clarify that farm taps (i.e. gas pressure regulation installed on services connected to
transmission pipelines) are not included in the count of metering and/or regulator stations.
D. Clarify Component Count Requirement
Proposed section 98.236(c)(19) Data Reporting Requirements states: For fugitive emission
sources using emission factors for estimating emissions report the following: (i) Component
count for each fugitive emission source. If reporting will be required by component count
instead of by station or facility, AGA recommends that EPA:
Clarify that only a count of leaking components by type are required to be reported, not a count
of all components by type; and
Establish a size threshold for components, such as 2 inch nominal diameter.
Response: EPA agrees that it should clarify the source category definition of natural gas
distribution to not include transmission pipeline. In todays final rule, the definition of natural
gas distribution in 98.230 explicitly states that natural gas distribution does not include interstate
and intrastate transmission pipelines.
1742
EPA disagrees that it must clarify that only the distribution side of gate stations are required for
reporting under the natural gas distribution facilities. Please see the response to EPA-HQ-OAR
2009-0923.
EPA agrees, and has revised todays final rule to clarify that customer meters are not required for
reporting. Please see the comment response in Section II.F of the preamble to todays final rule.
Additionally, the same section of the preamble also clarifies which metering and regulator
stations are required for leak detection and how to account for the remaining stations.
EPA agrees and todays final rule clarifies that farm taps are not included in the definition of the
natural gas distribution segment. Please see the response to EPA-HQ-OAR-2009-0923-1016-23.
EPA agrees that it is beneficial to clarify the data reporting requirements for equipment leaks
found in each leak survey. Todays final rule states that total count of leaks found in each
complete survey must be listed by the date of the survey and each type of leak source for which
there are leaker factors in Tables W-2, W-3, W-4, W-5, W-6, and W-7 of the rule.
EPA agrees that it should establish a threshold for component sizes required for leak detection;
however, it disagrees that the threshold should be 2 inches. Todays final rule stipulates that
tubing systems equal or less than one half inch diameters neednt be reported under the
requirements of subpart W for leak detection and leaker emission factors and population count
and emission factors, to be consistent with the leak detection and repair program (LDAR).

The Definition of Facility for Distribution is Overbroad
EPA is proposing to cast aside the normal, common-sense definition of facility in the case of
natural gas distribution. The definition used for decades in other Clean Air Act programs is the
same as the definition provided in Subpart A of the GHG Mandatory Reporting Rule (MRR).
The term facility is defined in 40 C.F.R. 98.6 as follows:
Facility means any physical property, plant, building, structure, source, or stationary equipment
located on one or more contiguous or adjacent properties in actual physical contact or separated
solely by a public roadway or other public right-of-way and under common ownership or
common control, that emits or may emit any greenhouse gas.
We support EPAs proposal to use this traditional definition of facility for natural gas pipeline
compression, underground storage, LNG storage, and LNG import/export facilities that our
members operate. However, instead of using the traditional definition of facility for natural gas
distribution, EPA is now proposing to define a gas distribution facility to encompass virtually
the entire distribution system operated by a single company, which can span an entire state.
1743
Proposed section 98.238 defines the term as follows:

independent municipally-owned distribution system. We would prefer a consistent, normal
definition of the term facility throughout the MRR limited to sites with contiguous
boundaries. This type of facility-level reporting by owners and operators will be consistent with
other Clean Air Act and state-level regulatory programs, which will facilitate compliance and
minimize the administrative burden of the proposed rule.
B. In the Alternative, EPA Should Clarify that the Unusual Definition of Facility for GHG
Reporting Purposes Should Not Be Used for the PSD Tailoring Rule or Other Clean Air Act
Purposes
While Consumers agrees that it will be useful to develop a good estimate of greenhouse gas
emissions from natural gas distribution, we do not agree that this requires such an expansive
definition of distribution facility. We are particularly concerned that this expansive definition
could be adopted in other contexts, such as future phases of the Prevention of Significant
Deterioration (PSD) Tailoring Rule for greenhouse gas emissions. If EPA does not adopt the
traditional definition of facility for natural gas distribution, then the agency should at least
clarify that other Clean Air Act programs should continue to use the traditional definition of
facility to determine whether emissions from a facility exceed regulatory thresholds requiring
pre construction permits or the application of control technology. Otherwise, far too many
operating decisions within a state-wide local distribution system could require time and resource
consuming regulatory decisions.
Response: EPA disagrees that it must redefine the natural gas distribution facility definition in
98.238 to adhere more closely to the definition of facility in subpart A. Please see the
The current definition includes the infrastructure spanning more than one state if it is part of the
same system. Please see the response to EPA-HQ-OAR-2009-0923-1016-26.
EPA disagrees that it should stipulate in subpart W that other section of the Clean Air Act should
not use this definition of natural gas distribution facility. Please see the response to EPA-HQ
OAR-2009-0923-1016-26.

Definition of a Natural Gas Distribution Facility:
Laclede does not believe that the entire distribution system is truly a Facility as intended by the
1744
Clean Air Act. Laclede believes that EPA incorrectly attempts to encompass a gas distribution
system into its concept of a facility for the purpose of aggregating CO2e emissions. The
distribution system piping network should not be used as a surrogate regulatory mechanism to
connect together any number of otherwise distinct gas company components, located at widely
dispersed locations, i.e. service buildings, compressor stations, process heaters, etc. By doing so,
EPA is grasping for the reporting of inconsequential, incremental emissions. This is a classic
case of diminishing returns for the investment. The gas distribution system should not be
viewed/defined by EPA in this manner because, if total fugitive emissions calculated using
EPAs approved emission factors for the gas distribution system exceed 25,000 metric tons/yr
CO2e, it should certainly not automatically trigger reporting of combustion emissions from the
LDCs other sites, when facilities at such sites were not otherwise required to report under
existing subpart C. This is an unreasonable stretch.
Response: EPA does not agree that the definition of an LDC facility in Section 98.238 of
todays final rule must conform with other Clean Air Act facility definitions or the facility
definition in Subpart A, Section V of the Final Rule, October 2009. For more information see
response to EPA-HQ-OAR-2009-0923-1016-26 and EPA-HQ-OAR-2009-0923-1044-1. EPA
agrees that small combustion emissions need not be reported, and has established an external
combustion equipment threshold. For more information on the threshold, please see the
rulemaking docket (EPA-HQ-OAR-2009-0923) under Equipment Threshold for Small
Combustion Units. Internal combustion engines which would have to be reported under Subpart
C for LDCs that exceed the 25,000 tCO2e reporting threshold must still be reported, but under
Subpart W starting data collection year 2011.

EPA apparently recognizes that its definition of a Facility on page 56387 of the October 30,
2009 final rule is incongruent with a natural gas distribution system.
The April 12, 2010, proposed rule includes, at 40 CFR 98.230 (8), Natural Gas Distribution,
under Definition of the source category. At yet another location in the proposed rule, specifically
at 40 CFR 98.238, Definitions, EPA goes on to provide an additional definition for Natural gas
distribution facility. This appears to be EPAs attempt to define the entire gas distribution
system as a Facility for purposes of the rule.
All of these definitions are yet in addition to the amended definitions on Page 18633 of the
proposed rule; 40 CFR 98.6, Definitions. Laclede questions why EPA finds it necessary to list
definitions at so many different locations throughout the rule. It appears this could be handled in
a more efficient and straightforward manner.
Response: Definitions under 40 CFR 98.6 are applicable to all the subparts under 40 CFR part
98. Definitions listed under 40 CFR 98.238 either override definitions under 40 CFR 98.6 or
1745
delineate items that are only applicable to Petroleum and Natural Gas Systems (40 CFR part 98,
subpart W). In todays final rule, the natural gas distribution source category is described in 40
CFR 98.230. The definition of a Facility for purposes of natural gas distribution is provided
under 40 CFR 98.238, which is only applicable to Petroleum and Natural Gas Systems (40 CFR
part 98, subpart W). It is necessary to distinguish natural gas distribution because the 98.6
definition is not appropriate to that sector and emissions which need to be reported from that
sector. See, EPA-HA-OAR-2009-0923-1016-26.

Metering & Regulation (M&R) Stations
DPL finds it is necessary for the EPA to clarify Subpart W's definition of above-grade M&R
stations and the extent to which M&R stations would be subject to leak-detection surveys.
Therefore, DPL requests that Subpart W clarify and/or prescribe certain M&R stations intended
for surveys based on M&R station function, purpose, size, and/or pressure. DPL also requests
that customer meters be excluded from leak detection surveys due to the extraordinarily high
number of such meters.
Response: In todays final rule, the definition of natural gas distribution has been clarified in
98.230 to state that customer meters are not included. For further information on this comment,
please see the response to comments in Section II.F of the preamble.
EPA has determined that requiring leak detection at meters and regulators based on size and/or
pressure is not necessary. In EPAs expert judgment, defining a size and pressure limit for meter
and regulator stations is difficult and confusing to reporters. As result, EPA decided to limit leak
detection to only above-grade meters and regulators at city gate stations with custody transfer.
For further information on this issue, please see the response to comments in Section II.F of the
preamble.

Based on the fact that the proposed rule does not provide clear definition of several key terms,
including "optical gas scanning", "M & R stations".
required for leak detection, please see the response to comments in Section II. F of the preamble.
In the proposed rule, the language optical gas scanning is not used. However, if the commenter
is referring to optical gas imaging, the EPA does not deem it necessary to define optical gas
1746
imaging instrument because procedures to use such an instrument and the necessary
specifications of the instrument are outlined under 40 CFR 98.234.

EPA must clearly define the term "M & R station" to only include industrial metering and
regulating stations.
required for leak detection, please see the comment response in Section II. F of the preamble.

Definition of Facility
The concept of facility as it applies to a local distribution company (LDC) is still undefined
within the proposed regulation. As indicated by AGA in its testimony in April, the proposed rule
would likely apply to many more gas utilities and require a much greater burden than estimated
by EPA. It is not clear whether EPA intends to define the facility as each distribution system or
all the distribution systems owned by a single Distribution Company. If the intent is the latter,
to combine all the separate distribution systems of a single company as one facility that spans
hundreds of communities across a state, then many more utility companies would likely exceed
the 25,000 ton per year threshold for reporting. Once included, those utilities would then have to
estimate and report emissions for many small meter and regulator stations and city gate stations
scattered across the state - as well as combustion emissions from sources that otherwise would
not exceed the facility reporting threshold. Many of the LDCs in the Northeast operate in various
locations without direct or contiguous links to each other; it is uncertain how this definition
applies
Response: EPA disagrees that the facility definition in 98.238 for natural gas distribution needs
to be clarified. Please see response to comment EPA-HQ-OAR-2009-0923-1016-26. EPA does
not agree that the definition of an LDC facility in 98.238 must conform with other Clean Air Act
facility definitions or the facility definition in Subpart A, Section V of the Final Rule, October
2009. For more information see response to EPA-HQ-OAR-2009-0923-1016-26 and EPA-HQ
OAR-2009-0923-1044-1.
1747

EPA CAA programs:
- 40 CFR Part 60 Subpart KKK, Standards of Performance for Equipment Leaks of VOC
from Onshore Natural Gas Processing Plants, defines Natural gas processing plant (gas
plant) as any processing site engaged in the extraction of natural gas liquids from field
gas, fractionation of mixed natural gas liquids to natural gas products, or both. Subpart
KKK also defines Natural gas liquids as the hydrocarbons, such as ethane, propane,
butane, and pentane that are extracted from field gas. As defined here, natural gas
liquids are those light-end hydrocarbons that are extracted from a natural gas stream that
must be stored in a closed pressurized tank to be maintained as a liquid. Taken together,
these definitions do not incorporate the removal of water, condensate, oil, H2S, CO2, or
any other component other than natural gas liquids.
- 40 CFR Part 68, Chemical Accident Prevention Provisions, defines Natural gas
processing plant (gas plant) as any processing site engaged in the extraction of natural gas
liquids from field gas, fractionation of mixed natural gas liquids to natural gas products,
or both, classified as North American Industrial Classification System (NAICS) code
211112 [previously Standard Industrial Classification (SIC) code 1321].
In addition, the oil and gas industry commonly understands the term natural gas processing
plant to mean the extraction of natural gas liquids from a natural gas stream to meet the heat
content specification of sales contracts. A Joule-Thompson unit or a refrigeration unit typically
completes this natural gas liquids extraction. It is not appropriate to introduce a different
definition for natural gas processing plant in this program. The contrast between industrys
understanding of the term and the Subpart W definition of the term will lead to confusion among
owners and operators, and will be inconsistent with the way EPA applies it under other parts of
the CAA. IPAMS requests that EPA define natural gas processing plant as a processing site
engaged in the extraction of natural gas liquids from field gas, fractionation of mixed natural gas
liquids to natural gas products, or both.
Response: EPA reviewed and considered other definitions used in the Clean Air Act and
implementing regulations, and where suitable for the purposes of subpart W, such as
transmission compressor stations, adopted those facility definitions. In the case of gas processing
facilities, EPA determined that limiting the definition to facilities that are significant sources of
criteria air pollutant emissions, such as gas fractionation facilities, and excluding those facilities
that have minimal criteria pollutants but significant greenhouse gas emissions is contrary to the
intent of the Mandatory Reporting Rule, and therefore, EPA disagrees with this commenters
request. Please see response to comment EPA-HQ-OAR-2009-0923-1044-1. However, EPA has
revised todays final rule for the source category definition of natural gas processing to include
1) all processing facilities that fractionate and 2) those that do not fractionate with throughput of
25 MMscf per day or greater. Please see Minimum Gas Processing Throughput in EPA-HQ
OAR-2009-0923. In addition, EPA has revised todays final rule and no longer includes
1748
gathering lines and boosting stations in natural gas processing. Please see the preamble Section
II.F.
75 Fed. Reg. at 18636 (emphasis added).
specification of sales contracts. A Joule-Thompson unit or a refrigeration unit typically
longstanding usage of the term and other EPA regulations with the proposed definition in
Subpart W will lead to confusion among owners and operators, and will be inconsistent with
other existing EPA regulations under the Clean Air Act.
Accordingly, Anadarko strongly recommends that the definition in Section 98.230(a)(3) be
revised to exclude field gathering and boosting stations, including any associated acid gas
treating facilities, from the Onshore Natural Gas Processing source category definition, and that
a new source category definition be included for Onshore Natural Gas Gathering Compression
and Treating Facilities. Using the definition of Onshore Natural Gas Transmission as a
template, Anadarko respectfully requests that the source category definition for Onshore
Natural Gas Gathering Compression and Treating Facilities be changed to the following:
Onshore natural gas gathering compression and treating facilities means any physically adjacent
combination of compressors that move natural gas from production fields or other compression
facilities into natural gas processing facilities, other gathering compression facilities,
transmission pipelines, storage facilities, or other end users. In addition, natural gas gathering
compression and treating facilities may include equipment for liquids separation, natural gas
dehydration, acid gas removal, and tanks for the storage of water and hydrocarbon liquids. These
facilities do not include equipment designed to extract natural gas liquids.
Response: EPA does not agree with this request to limit the gas processing facility definition to
that in the Clean Air Act. For more information, reference EPA-HQ-OAR-2009-0923-1298-25.
However, EPA has revised todays final rule and no longer includes gas gathering lines and
boosting stations. See Section II.F of the preamble.

1749
specification of sales contracts. A Joule-Thompson unit or a refrigeration unit typically
longstanding usage of the term and other EPA regulations with the proposed definition in
Subpart W will lead to confusion among owners and operators, and will be inconsistent with
other existing EPA regulations under the Clean Air Act.
Response: EPA does not agree with this request. For more information, reference EPA-HQ
OAR-2009-0923-1298-25.

inappropriate. The definition of a gas processing plant is well established in the context of the
CAA, and it does not incorporate stand-alone gathering system compression facilities. Gas
compression facilities, whether in the gathering systems upstream of gas processing or in the
natural gas transmission sector, are discreet and easily identifiable facilities. These facilities have
been treated as stand-alone sites in all previous applications of the CAA, including all permitting
programs and for GHG reporting under 40 CFR Part 98 Subpart C. Continuing this historical
Support Document for the Petroleum and Natural Gas Industry, which states: The term natural
gas processing plant has been previously and consistently defined in other EPA programs. A
natural gas processing plant (gas plant) is defined under Standards of Performance for
Equipment Leaks of VOC from Onshore Natural Gas Processing Plants as any processing site
engaged in the extraction of natural gas liquids from field gas, fractionation of mixed natural gas
liquids to natural gas products, or both. See 40 CFR Part 60, Subpart KKK. Subpart KKK also
defines natural gas liquids as the hydrocarbons, such as ethane, propane, butane, and pentane that
are extracted from field gas. Accordingly, natural gas liquids are those light-end hydrocarbons
that are extracted from a natural gas stream that must be stored in a closed pressurized tank to be
maintained as a liquid. Taken together, these definitions do not incorporate the removal of water,
condensate, oil, H2S, CO2, or any other component other than natural gas liquids.
1750
Response: Todays final rule does not include gathering lines and boosting stations and hence
the issue of gathering pipelines is not relevant. For further clarification, please see Section II.F of
the preamble. EPA does not agree with limiting gas processing to fractionation consistent with
40 CFR Part 60, Subpart KKK. For more information, reference EPA-HQ-OAR-2009-0923
1298-25 and EPA-HQ-OAR-2009-0923-1044-1. In the final rule EPA has revised the definition
of natural gas processing facilities in 98.230, to include those with fractionation, and those
without fractionation with a gas throughput capacity of 25 mmscfd and above. Please see the
preamble Section II.E and Minimum Gas Processing Throughput in EPA-HQ-OAR-2009
0923.
Organization:
The definition of a natural gas processing facility under Subpart W is inconsistent with the
definition of a natural gas processing facility under Subpart C, as Subpart W includes booster
stations and gathering lines, while Subpart C does not. In many situations, extensive field work
has already been completed and significant expenses incurred to assess the need for a natural gas
processing facility to report under Subpart C for the 2010 reporting year. The inclusion of
booster stations will create additional burden as companies will need to revisit every facility to
include booster stations. We propose that booster stations be viewed as separate facilities within
themselves
the Preamble.

The definition of a gas processing plant is well established in the context of the CAA, and it does
not incorporate stand-alone gathering system compression facilities. Gas compression facilities,
whether in the gathering systems upstream of gas processing or in the natural gas transmission
1751
sector, are discreet and easily identifiable facilities. These facilities have been treated as stand
alone sites in all previous applications of the CAA, including all permitting programs and for
greenhouse gas (GHG) reporting under 40 CFR Part 98 Subpart C. Continuing this historical
Support Document for the Petroleum and Natural Gas Industry, which states:
U.S. Environmental Protection Agency, Fugitive Emissions Reporting from the Petroleum and
Natural Gas Industry: Background Technical Support Document (TSD).
the preamble.

lend themselves to identifying scope of reporting and responsible reporting entities. 75 Fed. Reg.
at 18613.
EPA programs. A natural gas processing plant (gas plant) is defined under Standards of
Performance for Equipment Leaks of VOC from Onshore Natural Gas Processing Plants as any
processing site engaged in the extraction of natural gas liquids from field gas, fractionation of
mixed natural gas liquids to natural gas products, or both. See 40 C.F.R. Part 60, Subpart KKK.
Subpart KKK also defines natural gas liquids as the hydrocarbons, such as ethane, propane,
butane, and pentane that are extracted from field gas. Accordingly, natural gas liquids are
those light-end hydrocarbons that are extracted from a natural gas stream that must be stored in a
closed pressurized tank to be maintained as a liquid. Taken together, these definitions do not
incorporate the removal of water, condensate, oil, H2S, CO2, or any other component other than
natural gas liquids.
Additionally, natural gas processing plant (gas plant) is defined under the Chemical Accident
Prevention program as any processing site engaged in the extraction of natural gas liquids from
1752
field gas, fractionation of mixed natural gas liquids to natural gas products, or both, classified as
North American Industrial Classification System (NAICS) code 211112 [previously Standard
Industrial Classification (SIC) code 1321]. See 40 C.F.R. Part 68.
the processing plant fence, are considered part of natural gas processing plant. 75 Fed. Reg. at
18636 (emphasis added).
Response: EPA does not agree with this request to define gas processing facilities as
fractionation plants. For more information, reference EPA-HQ-OAR-2009-0923-1298-25.
Todays final rule does not include gathering lines and boosting stations and hence the issue of
gathering pipelines is not relevant. For further clarification, please see Section II.F of the
preamble.
Onshore Natural Gas Processing Plants Definition
The proposed onshore natural gas processing plants definition that combines GHG emissions
from gathering compressor facilities and the gathering pipelines with gas processing plant
emissions is in conflict with existing CAA regulations and inconsistent with industrys
longstanding usage of the term.
Major compression facilities are already subject to Subpart C reporting since their combustion
emissions exceed 25,000 metric tones per year. In addition, smaller compression facilities, such
as gathering compression facilities, with green house gas emission in excess of 25,000 metric
tones per year, from combined combustion (Subpart C) and methane (Subpart W) emissions
would also be required to report. The bulk of emissions from compressor facilities is from CO2
yet the aggregated definition of onshore natural gas processing plants in this subpart W
proposal will significantly increase reporting from very small compressor facilities that,
individually, would have relatively insignificant green house gas emissions, especially methane.
1753
In fact, the proposed Subpart W will require reporting of gas gathering and processing facilities
to nearly 100% for Devon, which is unreasonably burdensome. We strongly recommend that
EPA excludes field gathering and boosting stations from the onshore natural gas processing
plants source category definition.
the preamble.
The definition of a gas processing plant is well established in the context of the CAA. It does not
incorporate stand-alone gathering system compression facilities. Gas compression facilities are
discrete and easily identifiable facilities that have been treated as stand-alone sites in all previous
applications of the CAA, including all permitting programs and for GHG reporting under 40
CFR Part 98 Subpart C.
the preamble.

Reporting of field gas combustion emissions should not be accomplished by adopting an
excessively broad definition of pipeline quality natural gas, because the resulting use of Subpart
C Table C-1 default factors will often lead to excessive errors in estimating emissions.
Response: EPA agrees with the commenter and added methods for onshore production and
natural gas distribution to estimate the composition of field gas in lieu of using default factors in
Subpart C. In todays final rule, reporters must determine the composition of field gas using
either a continuous gas analyzer or use the most recent gas composition based on available
sample analysis of the field.

1754
Meter regulators: The current draft does not appear to include definitions for "gate stations" or
"meter regulators". While the term "gate stations" is relatively clear, what is intended for "meter
regulators" is not, and so it is difficult to gauge the full scale and impact of the draft rule's
reporting requirements. We suggest that EPA consider removing the requirement for leak
detection at these "meter regulators" and include them with the sources that are covered by
equipment count and population emission factors. We also suggest additional station definitions
be included in the rule.
Leak definition: EPA is specifying the use of optical gas imaging instruments for leak detection
at facilities. Yet it appears from the discussion in the EPA's Fugitive Emissions Reporting from
the Petroleum and Natural Gas Industry: Background Technical Support Document, p. 42, that
the definition of leaks in the context of this equipment requires some interpretation as to
detection sensitivity. We would like to see additional clarity on this in the final rule.
Response: EPA agrees, and has revised todays final rule to clarify under the natural gas
distribution source category, which M&R stations are required for leak detection. Please see the
comment response in Section II.F of the preamble. EPA has revised todays final rule for optical
gas imaging instruments in 98.234 that any emission detected by the optical gas imaging
instrument is a leak.
It is unclear what the EPA means by "continuous." It is unclear at what point is the unit
considered not in operation and therefore exempt. A general note regarding the treatment of high
and low-bleed devices: PAW recommends that EPA harmonize treatment of these sources with
the API Compendium.
Response: For further clarification on the meaning of continuous and on the methodology to
estimate emissions from these sources, please see the response to EPA-HQ-OAR-2009-0923
1060-28.

98.233(a)(1)(ii): Calculate the natural gas emissions for each continuous bleed device using
Equation W-1 of this section. EPA should define continuous.
Response: EPA agrees with the comment. For a clarification on the meaning of continuous.
1755

98.232(j): Flares. PAW believes that the EPA needs to clarify the definition of this emission
source. Are enclosed flares or fireboxes included in this source category?
Response: EPA has reviewed this comment and notes that the flare source in todays final rule
does not include enclosed flares (e.g. incinerators) or the firebox of enclosed thermal oxidation
units. These sources were deemed small and were not included in the rule. See Section 4.c. of the
TSD to todays final rule for further details on the decision process.

98.233 (n):
During the review of 98.233(n)(2)(i) CAPP noticed that the terminology is not representative of
most gas plants. The reference to de-methanizers is inaccurate as this process is typically found
on natural gas liquid recovery or fractionation plants and not gas plants. Taking this into account,
CAPP requests that EPA provides guidance on where the break point would be if there isnt a
demethanizer at the gas plant. One option CAPP proposes would be to use the "hydrocarbon
dewpoint control plant" as the break point.
Response: EPA agrees with the comment. In the absence of a de-methanizer in the gas
processing facility, todays final rule has been updated to require dewpoint control point as the
demarcation point.
98.233 (w):
For 98.233(w) CAPP recommends that the EPA replaces the term "EOR injection pump" with
"EOR injection system" since the use of compression or pumping is determined by the physical
properties of the CO2 being injected, which can be either gaseous or supercritical dense phase
which behalves like a liquid.
As with the other sections which refer to acceptable "consensus-based standards organizations"
1756
CAPP requests that the EPA provide clarity on 98.233(w)(3) equation W-24 definition for Rc
and provide a list of acceptable "consensus-based standards organizations".
Response: EPA does not agree on the use of the term EOR injection system. EPA requires the
monitoring of compressors in onshore production that deal with CO2 streams and hence the
concern about gaseous phase CO2 streams in EOR operations is unfounded. Hence, EPA has
retained the term EOR injection pump in todays final rule.
EPA does not understand what is unclear to the commenter about the definition of Rc. EPA has
retained the definition of Rc as the density of EOR injection gas in critical phase. With regard to
a list of consensus-based standards organizations, please see the response to EPA-HQ-OAR
2009-0923-1018-33.

98.236 (f):
In reviewing 98.236(f) CAPP noticed that the requirement to report emissions separately for
portable equipment for drilling rigs, dehydrators, compressors, electrical generators, steam
boilers, and heaters is not consistent with the portable equipment definition provided in the
requirements section of Sub-Part W. It is recommended that the EPA ensure that all sections
within Sub-Part W are using consistent definitions.
Response: EPA agrees with the comment. In todays final rule, EPA has moved the listing of
portable equipment into Section 98.232 and aligned it with data reporting requirements in
Section 98.236.

Commenter: Kim Dang
Reporting of Operational Data. The proposed rule also requires that minimum, maximum, and
average throughput be reported for each operation of the petroleum and natural gas sector for
which reporting is required under Subpart W. 262 It is not clear whether EPA intends for the term
operation to refer to an entire facility, or to individual compressors or other subunits of a
facility. Depending on which meaning EPA intends, this information could be problematic to
262
Proposed 40 C.F.R. SECTION 98.236(d).

1757
obtain. Just as important, the information is not useful or relevant to achieving the objectives of
the Mandatory Reporting Rule; in our experience, throughput levels bear little if any relationship
to the GHG emission profiles of our facilities. Accordingly, Kinder Morgan recommends that
EPA omit this requirement from the final rule.
Response: EPA agrees with the commenter in regard to minimum, maximum and average
throughput. Although, in todays final rule, EPA has clarified that only annual throughput using
engineering estimation based on best available data is required to be reported at a facility level;
the use of the term operation is inconsistent with the terminology of the rule. The throughput
levels are necessary for any policy making that may require a focused approach based on the size
of the facility. The commenter has not provided any data to substantiate the claim that emissions
profiles are not related to facility size. Although this could potentially be true for equipment
leaks, vented emissions could potentially be correlated with facility size. In the absence of any
data at a national level, EPA does not see an issue with collecting some basic information that is
already known to the reporter and would help in policy making.

Alternatively, if EPA determines that GHG emissions from gathering lines must be reported,
we recommend that a definition for gathering lines should be included in the rule. Such a
definition should specify that gathering lines: (1) are located upstream of natural gas
processing plants; (2) carry produced gas; (3) do not include piping within gas plants,
compression facilities, and treatment facilities; and (4) do not include transmission lines,
even if located upstream of a gas plant.
Response: In todays final rule, EPA has not included gathering lines and boosting stations as an
emissions source. For further information on this issue, please see Section II.F of the preamble to
todays final rule.

Why was the AAPG-CSD Geological Provinces Code Map selected? This is a document that
most operations and engineering professionals are not familiar with. Was the use of counties/
parishes considered or some other better known cartographic boundary considered
Response: The AAPG definition of basins uses county boundaries to demarcate a basin. Please
see Section II.D of the preamble to todays final rule for further details on why AAPG geologic
province code was the preferred definition for onshore production.
1758

Clarification is Needed on Vents Included under 98.233(p)
Vents that require measurement should be specified and clearly defined. It could be interpreted
that 98.233(p)(2)(i) reference to all vents means only those atmospheric vents associated with
reciprocating compressor rod packing, because that is the source type addressed by this section.
However 98.233(p)(2)(i) references rod packing, unit isolation valves, and blowdown valves.
INGAA requests that EPA specifically list the emission sources and associated vent lines to
clarify required measurements for each operating mode. For example, in the de-pressurized
mode, it is INGAAs understanding that leakage past unit isolation valves that find a path to
atmosphere through the blowdown vent is the emission source of concern to EPA. Additional
clarification and definitions should be provided to specify the emission sources and associated
vents.
Vented and fugitive emissions are better defined than in the 2009 Subpart W proposal. However,
consistent nomenclature is ultimately required to eliminate ambiguity, redundancy (i.e., double
counting) and unnecessary confusion in rule implementation. EPA should clearly define and
elaborate upon the intended source(s) and the measurement point(s) for reciprocating compressor
rod-packing vents under each operating mode. These nuances and measurement criteria are not
clearly defined and should be addressed in the Final Rule.
Finally, as discussed in Comment X, reporting individual compressor throughput is not feasible
or practical and should be deleted from the rule
Response: EPA agrees with the commenter on the combining of multiple venting sources under
the rod packing source type. EPA has revised the source type referred to as compressor vents.
Also, EPA now states what vent emissions are to be monitored in each of the different modes of
operations. With these changes EPA has determined there will not be any confusion over the
sources of venting emissions.
EPA does not agree with the commenter on the issue about individual compressor throughput.
EPA needs this information for any potential policy that wants to focus on a particular subgroup
of the population of sources. EPA is asking for an engineering estimation based on best available
data for this information. For further details, please see Section II.F of the preamble to todays
final rule.

1759

Lastly, we have concerns regarding the broad nature of the general definitions of terms as
established in previous rules. We believe these should more accurately define and represent the
industrys standard terminology.
Response: EPA understands the commenters concern and has addressed or outlined why EPA
does not agree with specific definitional concerns raised by commenters. For details on using
industrys standard terminology, please see the responses to EPA-HQ-OAR-2009-0923-1151-8
and EPA-HQ-OAR-2009-0923-1151-17.

Section 98.230(a)(1) Offshore petroleum and natural gas production. The rule text includes
inland waters which is not correct. In addition, Section 98.230(a)(1) and Section 98.238 both
provide the definition of offshore petroleum and natural gas production. The definitions differ in
that Section 98.238 states [a]ll production equipment that is connected via causeways or
walkways are one facility. This text should be added to Section 98.230(a)(1). API recommends
Section 98.230(a)(1) be revised to: Offshore petroleum and natural gas production is any
platform structure, affixed temporarily or permanently to offshore submerged lands, that houses
equipment to extract hydrocarbons from beneath the ocean or lake floor and that transfers such
hydrocarbons to hydrocarbon to storage, transport vessels, or onshore. In addition, offshore
production includes secondary platform structures and storage tanks connected to the platform
structure via causeways or walkways.
Response: EPA disagrees with the commenter on the exclusion of inland waters from offshore
and has retained it in the source category definition in Section 98.230 of todays final rule. EPA
agrees with the commenter regarding walkways, and has revised todays final rule offshore
source category definition in Section 98.230. EPA has also revised the definition of offshore in
Section 98.238, and clarifies that lakes are included.
Section 98.233(r)(2) Population count and emission factors. Coal Bed Methane Water Well
systems are not components, but instead are a type of oil production system. It would be more
appropriate to create a section to address these operations. In addition, it is too vague to say
"almost all CO2". A concentration should be provided.
1760
Response: Todays final rule does not include reporting of emissions from CBM produced
water. For further clarification on sampling of produced water, please see the response to EPA
HQ-OAR-2009-0923-1151-129 and Section II.E of the preamble to todays final rule.

Section 98.233(u)(2) GHG volumetric emissions. The terminology utilized in this section is not
representative of most gas plants, which do not utilize de-methanizers. De-methanizers are
typically utilized in natural gas liquid recover or fractionation plants. A more appropriate name
for use in place of De-methanizer would be hydrocarbon dewpoint control plant.
Response: EPA agrees with the comment. In the absence of a de-methanizer in the gas
processing facility, todays final rule has been updated to require dewpoint control point as the
demarcation point.
Section 98.238 Definitions.
Offshore petroleum, and natural gas production facility means each platform structure and all
associated equipment as defined in paragraph (a)(1) of this section. All production equipment
that is connected via causeway or walkways are one facility.
API assumes the reference to paragraph (a)(1) is referring to Section 98.230(a)(1).
Response: EPA agrees with the commenters on the incomplete reference to Section
98.230(a)(1). In todays final rule, the definition of offshore petroleum and natural gas
production facility is included in Section 98.230. Hence, the commenters issue is no longer
relevant due to the change.

Noble recommends that EPA conform the definitions and terminology used throughout the
proposed rule to be more consistent standard industry nomenclature to prevent errors and
confusion.
1761
- The rule should be consistent in defining standard conditions for expressing gas volumes and
mass and in its use of nomenclature; that is, references to industry conditions, ambient
conditions, actual conditions, and STP are confusing. Noble recommends that the rule
consistently apply 60 degrees F and 14.7 psia as the standard temperature and pressure; these are
the units commonly used in GHG reporting protocols and referenced industry standards, and are
the common units used for calibrating industry devices for custody transfer.
- 98.6 Definitions. Noble agrees with API comments concerning the definition regarding
Fugitive Emissions.
Response: EPA has revised todays final rule and uses the standard conditions of 68 degrees
Fahrenheit and 14.7 pounds per square inch absolute as defined in subpart A. For further details,
EPA disagrees with the proposed changes to the definition of fugitive emissions which are now
referred to as equipment leaks in todays final rule, please see the response to EPA-HQ-OAR
2009-0923-1151-17.
Noble requests clarification regarding definitions for flowlines and intra-facility gathering
lines referenced in Section 98.230(a)(2) with the clarification that an owner/operator is
responsible for reporting emissions from pipelines from the well, separator, compressor, etc. (as
applicable) to the point of custody transfer.
industry. Hence the commenter clarification is not required. For further details, please see
Section II.F of the preamble to todays final rule.

EPA Should Modify Reporting Requirements Applicable to Offshore Sources.
EPAs proposed definition of the Offshore petroleum and natural gas production source
category at 98.230(a)(1) should include all GHG emissions from offshore petroleum and
natural gas production, not just those limited to the time-period when the rig is attached to the
seabed floor. EPA should remove the phrase affixed temporarily or permanently to the offshore
submerged lands from the source category definition. All emissions associated with offshore oil
1762
and gas production must be reported, not just those that occur when the source is attached to the
sea floor.
EPA is proposing to amend 98.2(a) such that the reporting requirements apply to facilities
located in the United States or under or attached to the Outer Continental Shelf (as defined in 43
U.S.C. 1331), where the definition of United States includes territorial seas.
46 EPA states, [t]ogether these changes make clear that the Mandatory GHG Reporting Rule
applies to facilities on land, in the territorial seas, or on or under the Outer Continental Shelf, of
the United States, and that otherwise meet the applicability criteria of the rule. 263
We strongly support these definition changes in order to ensure comprehensive inclusion of
Outer Continental Shelf sources in the reporting requirements under Subpart W.
EPA should ensure that its GHG emissions reporting requirements from offshore oil and gas
sources are comprehensive and as thorough as those required for onshore sources. Likewise,
EPA should require offshore sources to use the same level of aggregation for determining
reporting requirements. For example, all of an operators offshore sources within an area that
corresponds to the area of a basin onshore should be required to aggregate emissions in order to
determine whether reporting requirements are triggered. Offshore oil and gas sources are
significant contributors to GHG emissions and must be held to the same standard as onshore
sources.
Response: EPA has determined that commenters assumptions and interpretations of the
proposed rule were not as intended by EPA and led to the view that drilling rigs not installed on
production platforms are reported under subpart W, or that other mobile sources of emissions
have to be reported. Subpart W applies predominantly to those production platforms reporting
activity date to the Bureau of Ocean Energy Management, Regulation and Enforcement
(BOEMRE) during a Gulfwide Emissions Inventory Study. This data is reported in the GOADS
software, and BOEMRE converts it to emissions and reports this data publicly. While BOEMRE
does research activity data on mobile sources (air and marine) as well as geogenic and biogenic
sources, and estimates the emissions from those, the data is not collected in a thorough survey of
every mobile source operator as platform data is collected. Therefore, todays final rule retains
the definition of platforms subject to reporting under this subpart to include those reporting
activity data in a GOADS survey, plus platforms in state and non-Gulf of Mexico Federal waters.
Finally, EPA clarifies that emissions sources not on platform facilities and mobile sources do not
have to be reported in todays final rule.

263
See 75 FR 18611, April 12, 2010.

1763
EPA should conduct a thorough investigation of geographic boundaries and organizational

boundaries (e.g., contractual relationships) in this industry sector to determine the best reporting
entity. As an example, CCAR is considering two options for defining a facility for the natural gas
transmission sector: (1) using the industry descriptions as portrayed in the North American
Industrial Classification System (NAICS); and (2) using the regulatory definitions from the
National Emission Standard for Emission of Hazardous Air Pollutants (NESHAPs) promulgated
by EPA 264 reporting rules, even if it is not feasible to include requirements in this version of the
rule.
Response: EPA reviewed both NAICS and NESHAP definitions in lieu of the Subpart W
definition for transmission segment and has determined that the Subpart W definition is robust
enough for the purposes of the MRR. Please see the responses to EPA-HQ-OAR-2009-0923
1024-14 for NAICS and EPA-HQ-OAR-2009-0923-3568.4-4 for NESHAP definitions.

This section requests minimum, maximum and average throughput for each operation listed in
paragraphs (a)(1) through (a)(8) of this section. This is a very vague request and has no
apparent meaning for most of the operations described in (a)(1) through (a)(8). In addition, this
information does not appear to be useful in determining or understanding the GHG emissions
information that is otherwise reported. Therefore, we recommend that paragraph (d) should be
deleted.

Section 98.2: Who Must Report The amended text of the proposed rule under the heading Who
must report? in Section 98.2(a) states:
The GHG reporting requirements and related monitoring, recordkeeping, and reporting
requirements of this part apply to the owners and operators of any facility [emphasis added]
It is IPAMS expectation that either the owner or operator would report emissions, not both
264
Discussion Paper for a Natural Gas Transmission and Distribution Greenhouse Gas Reporting Protocol, Final
Draft Report, Prepared for the California Climate Action Registry and World Resources Institute by URS
Corporation and The LEVON Group, June 6, 2007, pp. 4-20.
1764
parties, and that EPA did not intend for emissions to be double counted. Moreover, due to the
nature of the oil and gas production industry, there may be numerous owners of a facility that do
not have day-to-day control of a facilitys operations. Therefore, IPAMS requests that EPA
clarify Section 98.2(a) to state that the requirements of this part apply only to the operator since
this is the party that has control over a facilitys operations and therefore its GHG emissions.
Response: Owners or operators who are subject to reporting under todays final rule must
appoint a designated representative and only one designated representative who is responsible for
submitting reports. Please see Section 98.4 of todays final rule. It is the responsibility of
owners or operators to select that designated representative for reporting purposes as provided in
Section 98.4. The designated representative (DR) is the entity that is responsible for submitting
the emissions data pursuant to todays final rule. Please see the response to EPA-HQ-OAR-2009
0923-1024-16.

EPA should use the established CAA definition of a facility.
Section 98.230(a)(4) Onshore Natural Gas Transmission Compression
It is not clear where onshore natural gas transmission compression begins and modifying the
description to be consistent with the definition of this segment in Subpart HHH of the NESHAP
regulations is recommended.
D. Definitions and Terminology
BP requests that EPA conform the definitions and terminology used throughout the proposed
rule to make it more consistent with that used by industry. An example of the use of terms that
are not those commonly used by industry or equipment vendors, is the rules use of high bleed
devices rather than high bleed controllers, which is the term recognizable to equipment
vendors.
However, even more fundamental than the terminology issues, is the proposed rules confusing
1765
mix of conditions for expressing gas volumes and mass. The MRR uses, in different parts of the
existing rule (Subparts C, Y, NN, PP) as well as the proposed Subpart W, industry conditions,
ambient conditions, actual conditions and STP. EPA should consistently apply industry standard
conditions (60 degrees F and 14.7 psia) throughout the MRR since these are the units used in all
the referenced industry standards and are the common units used for calibrating industry devices
for custody transfer.
In addition, applying the emission equations in the proposed Subpart W produces a mix of
emission results. For example, Equation W-1 results in natural gas emissions (scf/yr) from high
bleed pneumatic devices, while Equation W-2 results in annual metric tonnes of CO2 equivalent
emissions from low bleed pneumatic devices. Although, 98.233(u) Equation W-22 converts
natural gas volumetric emissions to GHG volumetric emissions and 98.233(v) Equation W-23
converts GHG volumetric emissions to GHG mass emissions as CO2e, the mix of units resulting
for individual source types impedes comparison among sources. For consistency with existing
MRR reporting requirements and for comparison among emission source types, all emission
equations should result in metric tonnes of GHG emissions by GHG type (e.g., CO2, CH4, and
N2O) and then apply Equation A-1 to convert to CO2e emissions. Consistent use of industry
standard conditions for all calculations with conversion of volumes to metric tonnes for each
quantified source will ensure that annual GHG emissions are properly quantified and reported.
Response: EPA has reviewed this comment. EPA reviewed both NAICS and NESHAP
definitions in lieu of the Subpart W definition for transmission segment and has determined that
the Subpart W definition is robust enough for the purposes of the MRR. Please see the responses
to EPA-HQ-OAR-2009-0923-1024-14 for NAICS and EPA-HQ-OAR-2009-0923-3568.4-4 for
NESHAP definitions.
EPA has revised todays final rule and uses the standard conditions of 68 degrees Fahrenheit and
14.7 pounds per square inch absolute as defined in subpart A. For further details, please see the
EPA disagrees with the comment about using standard industry terminology and the consistency
of the equations with existing MRR reporting requirements. For further details, please see the
Equations W-20 and W-21 refer to ambient pressure. DTE Energy requests that EPA clarify
what is meant by ambient pressure to avoid confusion.
Response: EPA has reviewed the comment. However, in todays final rule EPA uses actual
conditions for equations W-20 and W-21. Since conditions at which emissions occur or are
monitored are not always the same as ambient conditions, the term actual conditions better
represent the conditions whether operational or ambient.
1766

On Subpart W, API requests that EPA conform the definitions and terminology used throughout
the proposed rule to make them more consistent to those used by industry. This is particularly
applicable to the rules confusing mix of expressions used for gas volumes and mass. For
example, the proposed rule uses in its different sections, industry conditions, ambient conditions,
actual conditions and STP. API continues to recommend that the rule should stick with, quote,
industry standard conditions (60F and 14.7psia) since these are the units used in all the
referenced industry standards and are the common units used for calibrating industry devices for
custody transfer.
Response: EPA has ensured that the relevant condition specifications are used for each of the
Subparts as required. Within Subpart W, EPA has reviewed and revised the temperature and
pressure conditions to specify either actual or STP as appropriate. For example, certain emissions
such as blowdowns are calculated at actual conditions and finally need to be converted to STP
conditions for reporting to EPA. Here the use of both actual and STP conditions is required in the
rule. Finally, emissions calculated in actual conditions are converted and reported to the EPA in
standard temperature and pressure. For further details, please see the response to EPA-HQ-OAR
2009-0923-1151-8.
1767

2010 Subpart-W RTC Part2

Uploaded by

Copyright:

Available Formats

2010 Subpart-W RTC Part2

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2010 Subpart-W RTC Part2

Uploaded by

Copyright:

Available Formats

VOLUME 8: COMPLIANCE AND ENFORCEMENT

COMPLIANCE AND ENFORCEMENT

Comment Number: EPA-HQ-OAR-2009-0923-1011-4

Preamble Page 18612

EPA Inventory of U.S. Greenhouse Gas Emissions and Sinks 1990-2008.

applicability of smaller facilities.

C. Safe harbor provision for screening methods:

Comment Number: EPA-HQ-OAR-2009-0923-1167-36

Comment Number: EPA-HQ-OAR-2009-0923-1082-8

Comment Number: EPA-HQ-OAR-2009-0923-1155-9

Comment Number: EPA-HQ-OAR-2009-0923-1045-9

VOLUME 9: LEGAL ISSUES

Comment Number: EPA-HQ-OAR-2009-0923-1024-16

Proposed Subpart W, 75 Fed. Reg. at 18,614.

Comment Number: EPA-HQ-OAR-2009-0923-1154-4

Comment Number: EPA-HQ-OAR-2009-0923-1154-18

Comment Number: EPA-HQ-OAR-2009-0923-1154-21

9.1.1 SEGREGATE OFFSHORE ISSUES SEPARATELY

Comment Number: EPA-HQ-OAR-2009-0923-1155-13

Federal Register Vol. 75, No. 69, 18611.

Federal Register Vol. 75, No. 69, 18611.

See 75 Fed. Reg. at 18,611.

42 U.S.C. 7414(a) (emphasis added).

Lockhart v. United States, 546 U.S. 142, 149 (2005).

See 43 U.S.C. 1333(a)(1).

See 43 U.S.C. 1332(3).

See 43 U.S.C. 1334(a).

CONFIDENTIAL BUSINESS INFORMATION (CBI)

Comment Number: EPA-HQ-OAR-2009-0923-1015-27

ownership structures (including percentage shares of owners, subsidiaries, holding companies,

Comment Number: EPA-HQ-OAR-2009-0923-1039-23

that are not emissions data, and are CBI.

Comment Number: EPA-HQ-OAR-2009-0923-1060-15

RELATIONSHIP TO OTHER CLEAN AIR ACT (CAA) PROGRAMS

Comment Number: EPA-HQ-OAR-2009-0923-2345-3

Comment Number: EPA-HQ-OAR-2009-0923-1044-1

Comment Number: EPA-HQ-OAR-2009-0923-1173-14

The Climate Registry.

COMMENTS RELATED TO PROCEDURES

Comment Number: EPA-HQ-OAR-2009-0923-1027-2

Comment Number: EPA-HQ-OAR-2009-0923-1173-6

COMMENTS ON PORTABLE EQUIPMENT REPORTING REQUIREMENT

Comment Number: EPA-HQ-OAR-2009-0923-3524-3

Comment Number: EPA-HQ-OAR-2009-0923-0582-8

Response: Please see the response to EPA-HQ-OAR-2009-0923-1040-5 for more details on

reporting requirements for portable equipment.

details, please see the response to EPA-HQ-OAR-2009-0923-1170-7.

Comment Number: EPA-HQ-OAR-2009-0923-1031-23

Commenter: Lisa Beal

See paragraph one under comments listed under 98.230 above.

Comment Number: EPA-HQ-OAR-2009-0923-1174-7

Portable Equipment Combustion Emissions

COMMENTS ON LEASED/RENTED EQUIPMENT REPORTING

Comment Number: EPA-HQ-OAR-2009-0923-1004-15

Comment Number: EPA-HQ-OAR-2009-0923-3568.4-5

Comment Excerpt Text:

COMMENTS ON REPORTING OF CONTRACTOR EMISSIONS

Comment Number: EPA-HQ-OAR-2009-0923-1040-5

Comment Excerpt Text: